University of Calgary PRISM: University of Calgary's Digital Repository

Graduate Studies Legacy Theses

2001 Populism and federalism: the interplay of direct democracy and federal institutions in Australia, Canada, Switzerland and the United States

Kamena, Theodore Henry Jr.

Kamena, T. H. (2001). Populism and federalism: the interplay of direct democracy and federal institutions in Australia, Canada, Switzerland and the United States (Unpublished doctoral thesis). University of Calgary, Calgary, AB. doi:10.11575/PRISM/16329 http://hdl.handle.net/1880/41090 doctoral thesis

University of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission. Downloaded from PRISM: https://prism.ucalgary.ca UNIVERSITY OF CALGARY

Populism and federalism: the interplay of direct democracy and federal

institutions in Australia, Canada, Switzerland and the United States

Theodore Henry Karnena, Jr.

A THESIS

SUBMITTED TO THE FACULTY OF GRADUATE STUDIES

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE

DEGREE OF DOCTOR OF PHILOSOPHY

DEPARTMENT OF POLITICAL SCIENCE

CALGARY, ALBERTA

APRIL, 2001

O Theodore Henry Karnena, Jr. 2001 National Libraty Bibliothhque nationale 1*1 of Canada du Canada Acquisitions and Acquisitions et Bibliographic Services services bibliographiques 395 Wellington Street 395, rue Wellington Ottawa ON KIA ON4 Ottawa ON K1 A ON4 Canada Canada Your lve Votre nllBme

Our fik Notre nlltimllcd

The author has granted a non- L'autew a accorde une licence non exclusive licence allowing the exclusive pennettant a la National Library of Canada to Bibliothkque nationale du Canada de reproduce, loan, distribute or sell reproduire, prSter, distribuer ou copies of this thesis in microform, vendre des copies de cette these sous paper or electronic formats. la forme de microfiche/^, de reproduction sur papier ou sur format electronique.

The author retains ownership of the L'autew conserve la propriete du copyright in this thesis. Neither the droit d'auteur qui protiige cette these. thesis nor substantial extracts fiom it Ni la th6se ni des extraits substantiels may be printed or otherwise de celle-ci ne doivent Stre imprimes reproduced without the author's ou autrement reproduits sans son permission. autorisation. ABSTRACT

This study examines the relationship between populism and federalism. In

theory, these two conceptions of society are in conflict. Populism, in its most elemental

form, sees the state as a tool for the "will of the people." This will is seen as indivisible,

indicating populism views society in a monist way. Federalism, on the other hand,

recognizes diversity within the state through the division of sovereignty between the

federal-level government and component parts of the state. Despite this theoretical conflict, however, populism, as operationalized by direct democracy, and federalism do co-exist in some federal states.

To examine this co-existence, this study uses data from ballot measures in four federal states - Australia, Canada, Switzerland, and the United States - using federal- level cases fi-om Australia, Canada and Switzerland as well as sub-national cases f?om selected Swiss cantons and American states in the period 1970-94. It also searches for intervening variables which may impact the relationship, focusing on the size of jurisdictions and the role of political parties.

The study concludes that populist devices may co-exist with federalism by serving as a check on national power, both through providing a popular check on legislative decisions and by restricting centralization in a federal system. Populism and federalism may also work together to provide different definitions of the "people" on different issues, but this is limited by the emergence and growth of both cooperative federalism and universal standards of rights. ACKNOWLEDGEMENTS

This project has been a part of my life for nearly seven years. I am grateful for the help and assistance of many people, I name but a few of them here.

During my 17-day visit to Switzerland in 1996, I was helped by many people. I am appreciative for the assistance of the staff of the Chancellorie dYEtatde Neuchatel, and the Staatskanslei in Kanton Luzern. In particular, I wish to thank Josef Koch of the

Staatskanslei for his patience in the face of a language barrier, and Alexandre Treschel of the Centre for the Study of Direct Democracy for his explanation of the einzelinitiative in

Zurich. I also wish to thank the University of Calgary for the travel grant which made the visit possible.

I am grateful as well to Campbell Sharman and Joel D. Sherman for their electronic correspondences, to Fred Wall for his assistance in gathering data from the

Angus Reid Group, to Came Sullivan for three invaluable months of aid as my research assistant, and Jan-Peter Schernrnel for his translation of much of the German language material.

I am thankful to the members of my committee, whose suggestions have vastly improved this work. And without Roger Gibbins, my Ph.D, advisor, I never would have reached this point.

I want to thank everyone in the political science department at Western

Washington University. It shall forever be my home, no matter where life may take me.

Thanks in particular to Gene Hogan for his incredible flexibility in finding ways for me to serve; to Todd Donovan, for being an incredible resource and sounding board on direct democracy; and most of all to Don Alper, who provided me avenues for pursuing my interest in Canadian politics f?om my first term as an undergraduate.

Finally, thanks to my wife, Christie and sons, Trevor and Derek. Christie, you have survived seven years of this, two of them far from other loved ones. I love all three of you more every day. TABLE OF CONTENTS

CHAPTER 1 .INTRODUCTION ...... 1 Questions to be raised ...... 2 Selection of data ...... 3 CHAPTER 2 - POPULISM ...... 13 What is populism? ...... 13 Rousseau and the "general will" ...... 18 The virtue of the "common people" ...... 20 The critique of representative democracy ...... 24 Populist support of direct democracy ...... 28 The use of direct democracy ...... 32 Differences in populist nature of referendums & initiatives ..37 Conclusion ...... -38 CHAPTER 3 - FEDERALISM ...... 42 What is federalism? ...... 42 Why federalism? ...... -48 A federalist conception of society ...... 54 Cooperative federalism ...... 59 Why federalism is worth studying ...... 62 CHAPTER 4 - POPULISM AND FEDERALISM ...... 66 Madison's critique of populism ...... 67 The shared liberty of federalism and populism ...... 70 The co-existence of federalism and populism ...... 72 CHAPTER 5 - POPULISM VERSUS FEDERALISM ...... 90 Cooperative federalism ...... -90 Populism's problem with cooperative federalism ...... 91 Canada ...... 95 United States ...... 97 Australia ...... 99 Switzerland ...... 101 Minority rights ...... 105 The issue of size ...... 110 Conclusion ...... 114 CHAPTER 6 - NATIONAL DISTINCTIVENESS AND CASE STUDIES ...... 117 Switzerland ...... 117 Federalism in Switzerland ...... 121 Direct democracy ...... 124 Australia ...... 132 Introduction of the constitutional referendum ...... 133 Referendums and Commonwealth power ...... 134 Political parties and the referendum ...... 137 Australia and Switzerland compared ...... -139 Canada ...... -141 The 1942 conscription plebiscite ...... 142 Charlottetown Accord ...... 145 Role of the parties ...... 149 Decentralization ...... 151 Complexity ...... 152 United States ...... -154 Minorities ...... -154 The upward movement of issues ...... 157 Conclusions ...... -161 CHAPTER 7 - CONCLUSION ...... 165 The questions raised ...... -165 Toward a "populist federalism"? ...... 170 BIBLIOGRAPHY ...... 173 APPENDIX 1 ...... 180 APPENDIX 2 ...... 184

vii CHAPTER ONE - INTRODUCTION

In the 1990s, the Reform Party of Canada made a dramatic rise to prominence,

going fkom obscurity to Official Opposition in a handhl of years. Among the slogans

Reform embraced in its early years were these two - "Common Sense of the Common

People," and "The West Wants In." The first slogan has been, in various forms, a

familiar one with populist movements across a wide range of political jurisdictions. The

second was a cry for representation, specifically, better representation within federal

institutions. The first is a call for power to the people, the second a call for power based

on and for the jurisdiction in which they live.

The two slogans are based in two differing and perhaps conflicting conceptions of

the state. A populist conception of the state, in its most elemental form, sees the state as

being a tool for the "will of the people." Populism sees this will as being indivisible, it

sees society in a monist way. All that is needed is to determine "common sense"

solutions. A federal conception, on the other hand, recognizes diversity within the state through the division of sovereignty between the federal-level government and the component parts of the state - provinces, states or cantons. These sub-national jurisdictions have differing populations and their presence could be seen as working against populism in that the "will of the people" may have difficulty being implemented across these jurisdictions.

The central conceptual point of this study is to examine the relationship of populism and federalism. If populism and federalism are theoretical enemies, how can there be coexistence? This is not questioning whether this coexistence exists, as it clearly does, but rather to ask what conditions this coexistence occurs under and what the nature

of the relationship is.

To determine this, we must be very specific about populism. The term is a multi-

faceted one, but one important demand of populist movements in democratic societies

has been for more direct democracy. Through direct democracy, and in particular, the referendum and initiative devices, the "will of the people" can be determined more precisely than it can through representative institutions.

Operationally, then, this study focuses on the use of direct democracy devices in federal states. More specifically, it raises a series of research questions.

QUESTIONS TO BE RAISED

I) What is the relationship of direct democracy to the operation of federalism?

This may sound like a broad question. It focuses on, but is not limited to, the balance of power between national and sub-national governments. Does direct democracy have any centralizing or decentralizing tendencies? Further to this, the impact of direct democracy on the negotiating that goes on between governments in federal states will be investigated.

2) What is the impact of direct democracy on minorities, in particular minorities that find a voice through federalism ?

A concern often raised regarding direct democracy is the danger of voters using a tyranny of the majority to take action against minority groups. This issue is particularly important in federal states, as one justification for federalism is to give local minorities jurisdictions where they can govern, or to preserve such jurisdictions. 3) What are the intervening variables that impact the relationship?

This question will be left open-ended in an effort to allow trends to emerge through the comparison presented. Nonetheless, specific focus shall be placed on two sets of variables. First, how do the size and diversity of jurisdictions impact the relationship? Do smaller, more homogeneous jurisdictions use direct democracy more, or less, and how is it used there? Secondly, what are the impacts of political parties and other institutions on the relationship between direct democracy and federalism?

4) How are the 'jpeople"defined?

This question is most closely linked to the central conceptual question to which all other questions are directed. Populism assumes the people can be defined and their will can be determined and carried out. Federalism assumes a divided society in which power is dispersed. How, if at all, are these two viewpoints reconciled?

SELECTION OF DATA

To consider these questions more thoroughly, this study shall examine the use of direct democracy in four federal states - Australia, Canada, Switzerland, and the United

States. From these four countries, a large data set of more than 2,000 ballot measures has been gathered. This data set includes the following fiom each country:

m Australia: All national referendums since the founding of the Commonwealth in

1901.

Canada: All national referendums and a selection of important provincial

referendums since and including the Conscription plebiscite in 1942. 4

a Switzerland: All federal ballot measures in the period 1970-1 994. Additionally,

the ballot measures from five cantons - Fribourg, Geneva, Luzern, Neuchatel, and

Zurich for the same period will be included.

United States: All ballot measures in the period 1970-1994 in five states -

California, Montana, North Dakota, Oregon and Washington.

For each ballot measure, the following information has been gathered - the subject of the measure, the type of measure (obligatory referendum, popular referendum, initiative and counter-proposals), raw numbers and percentages of yes and no votes, percentage of turnout, and, where available, the number of those who were at the polling booth but declined to vote on the measure. For federal level ballot measures, the number of sub- units approving the measurk will also be provided.

This data set was selected with a number of considerations in mind. First, the key was to find a broadly representative sample of the use of direct democracy in federal systems. Three of the countries were chosen in light of the frequency with which they use direct democracy. Switzerland uses ballot measures extensively both at the federal level and at lower levels. Australia uses referendums for the ratification of constitutional amendments, and is the second most frequent user, after Switzerland, of ballot measures at the federal (or national) level in the world. The United States was chosen because of the extensive use of ballot measures in the states, in particular the use of initiatives.

These state-level ballot measures provide the only truly useful comparison with the Swiss cantons, a comparison that is necessary to examine the impact of sub-national direct democracy on federal governance. 5 Within these three countries, fbrther considerations were taken into account in selecting the specific cases to be studied. For Australia, state-level referendums were not included because the issues have tended to be narrow in scope, and information on some of these measures is less readily available.

In Switzerland, concerns existed about the availability of data fiom very small cantons, including eight cantons having populations of less than 100,000. It was decided instead to concentrate on a selection of cantons which could provide a reasonable reflection of Switzerland's linguistic and religious diversity. From these, five cantons were chosen.

Zurich was selected as the largest canton and one in which direct democracy is used more frequently than in most cantons. It has a German-speaking majority and is historically Protestant, though only by a plurality today. While it contains Zurich,

Switzerland's largest city, only about 30 percent of the population of the canton lives in the metro area. As a counter-case, Geneva was selected as an important French-speaking canton. It is overwhelmingly urban, making it unique among the five selections.

Luzern was selected as a slightly smaller German Catholic canton. Thus it historically has been part of the linguistic majority in Switzerland, but the religious minority. As a counter-case to Luzern, Neuchatel was selected as a smaller French

Protestant canton. At a population of 165,000, it is the smallest of the five cantons in the study and ranks 16th in population among the 26 Swiss cantons. The largest cities in the canton, Neuchatel and La-Choix-de-Fonds, have populations of slightly more and slightly less than 40,000 people, respectively. 6 Fribourg was selected as one of only three Swiss cantons with more than one official language. It was selected rather than Bern. While having French as a second official language, Bern is largely ~erma&s~eaking,and Mer,is the second-largest canton in population. In these respects, it is much like Zurich, and thus was rejected.

Fribourg, where French is the majority language by roughly a 3-to-1 ratio over German, is 12th among the cantons in population.

Initially, concerns about availability of data proved to be reasonable. On-site research in Switzerland in May, 1996, indicated referendum data fiom Luzern and

Neuchatel could be gathered only through copying materials in cantonal administrative offices. More recently, data for ballot measures in all Swiss cantons have been compiled and placed on the internet by the Centre for the Study of Direct Democracy at the

University of ~eneva.'

In the U.S.,roughly half the states use ballot measures on at least a somewhat regular basis. The states were chosen, as with the Swiss cantons, for reasons of both theoretical significance and practicality. Montana was selected largely for reasons of research access, but also because it used somewhat fewer ballot measures than the other states selected. Also, pairing Montana with North Dakota and pairing Washington with

Oregon provides some basis for using a "similar-systems" approach (Lijphart, 1975) of comparison. California is the largest U.S.state and used more ballot measures in the period studied than any other state. These factors provide at least some controls for a comparison of California and Zurich, and the role influential sub-national jurisdictions many have on federal policies. 7 Further, the five states provide theoretical depth in two other respects. All five use ballot measures to varying degrees - California using it the most, Montana the least.

They have differing regulations on how to get initiatives to the ballot, which has an effect on the fiequency of use. Second, the five states selected provide a wide range of population and diversity. California is highly cosmopolitan, Oregon and Washington are becoming increasingly so. Montana and North Dakota, on the other hand, are more homogeneous and also less urban.

The inclusion of Canada was done on somewhat different grounds fkom the other three states. The fiequency of direct democracy has been extremely low compared to the other four countries. Furthennore, until the 1990s, and even then only on a very small scale, neither the federal government nor the provinces had any sort of legislation on obligatory referendums or popular initiatives, each of which can force measures onto a ballot. The primary justification for including Canada is not for the fkequency of ballot measures, but rather the importance of a handful of those measures, most importantly the

Charlottetown Accord referendum of 1992 and the two Quebec sovereignty referendums of 1980 and 1995.

A second justification also exists for including Canada. The 1990s witnessed initial efforts at obligatory referendums and popular initiatives. Alberta and British

Columbia have passed legislation requiring referendums prior to provincial passage of any constitutional amendment, and British Columbia and Saskatchewan now have guidelines in place for the use of popular initiatives, albeit guidelines far more restrictive than those in the United States or Switzerland. Alberta has also passed legislation requiring voter approval via referendum for any tax increase. Even where legislation is 8 not on the books, pressure for more direct participation in at least some decisions appears

to be growing. The Canadian Alliance, the successor party to the Reform Party, retains

as part of its platform a call for mandatory referendums on constitutional changes and the

introduction of the initiative device (Canadian Alliance, 2000). In the light of the reaction

to the closed process of the Meech Lake Accord and the eventual popular vote on the

Charlottetown Accord, some observers have claimed that future constitutional negotiations in Canada will have to be approved by the people (Gibbins & Thomas, 1993:

3). In light of the above developments, a greater understanding of the relationship between populism and federalism seems important to Canada. The interplay between the two appears to be increasing, even if not to the extent of the other three countries. Thus, while Canada fits in this study for slightly different reasons than do the other three countries, it does fit.

In addition to the introductory and concluding chapters, this study will include five substantive chapters. Chapters two and three serve as literature reviews of populism and federalism, respectively. Chapter two begins with a discussion of the theoretical basis of populism, focusing on discussions of the "general will," the populist belief in the virtue of the "common people," and the critique of representative democracy. This critique leads to populist advocacy of direct democracy, which is discussed in greater detail.

Chapter three focuses on federalism. Because of the wide range of research in federalism, the focus is somewhat narrower than chapter two. Initially, various definitions are discussed and attention is given to understandings of federalism focusing on divided sovereignty. From there, the chapter moves into a discussion of how federalism defines and divides the "people."

Chapter four brings the two concepts together. It discusses possibilities of populism and federalism coexisting and how this might occur. From that point, the empirical data are introduced, demonstrating that direct democracy and federalism have been able to coexist and what some of the impacts of that have been.

Chapter five seeks to examine the limits of the coexistence presented in chapter four. In particular, it focuses on three issues - the problems of cooperative federalism for coexistence, the rights of minorities in jurisdictions using direct democracy, and the issue of size. In the Federalist Papers, James Madison argued liberty was protected in larger, more diverse jurisdictions. This chapter suggests that populism argues in reverse of

Madison for smaller, more homogeneous federal sub-units.

Chapter six looks at the special characteristics of each of the states in this study.

In particular, the consensual nature of Swiss politics, the role of political parties in

Australia and the impacts of direct democracy on Canada will be discussed. Chapter six seeks to accomplish two goals - to find intervening variables in the relationship between populism and federalism and also to see what common threads may run through the cases.

Such a cross-national approach to the use of direct democracy has been unusual.

Butler and Ranney (1978 & 1984) have edited two books on the use of direct democracy around the world. However, their work is in the nature of a country-by-country or region-by-region analysis with little crossover between the independently written chapters. A wide-ranging search of bibliographies of major works on direct democracy 10 produced but one work which the title indicated was cross-national in nature. This was

an unpublished Ph.D. dissertation completed in 1977 comparing voting behavior in Ohio,

Oregon and ~witzerland.' Kobachls (1993) work on direct democracy in Switzerland

does include a chapter on other jurisdictions using direct democracy, including

information on Australia and California, but the extent of discussion is limited.

Information on direct democracy within each country is somewhat more

extensive. Regarding Switzerland, the major difficulty is finding works in English.

Some French literature is available, but the bulk of the material is in German. Two recent

English works, however, have provided detailed examinations of the Swiss political

system. Kobach (1993) focuses primarily on direct democracy and is cited in a wide

range of material, including by some Swiss scholars. A more general work on the Swiss

political system has been produced by Linder (1994). Linderls work proves useful

because of its detailed discussions of Swiss federalism, as well as direct democracy.

One examination of the relationship of direct democracy and federalism is provided by

the journal Publius (1989), which produced a special issue on Swiss democracy.

The other country with extensive research available on direct democracy is the

United States. A wide range of literature has been produced within the last two decades,

with Cronin (1989) providing perhaps the best blend of both the background and the

current use of direct democracy. (For an overview of present research, see Bowler &

Donovan, 1998.) A key issue in American research has been the impact of ballot

measures on minorites (Gamble, 1997; Magleby, 1984; Witt & McCorkle, 1997). Little of this work has directly examined federalism, but Magleby (1998) has investigated the 11 impact of initiatives as an agenda-setting device not only at the level they occur, but also the impact those initiatives have on federal policies.

Canada is also witnessing an increased amount of research into direct democracy.

Boyer (1993a&b, 1988) has been the most prolific, although his work is of a more polemical nature. The work of the Canadian National Election Study team on the 1992

Charlottetown Accord referendum (Johnston, et al, 1996) provides a detailed case study of that critical event. While Canada's experience with direct democracy is limited, the infomation on it is not all that different from the amount available fiom Australia.

Compared to the other three states, work on Australian referendums has been rather limited. As an example, an edited book on the 1977 Australian general elections

(Pennirnan, 1978) paid virtually no attention to the referendums held that year.

Furthermore, work on Australia has received relatively limited circulation in North

America. Again, a Publius (1990) special issue serves as an excellent gateway to other research.

Questions about the relationship between populism and federalism have been raised, at least indirectly, since Madison. That direct democracy devices exist within federal states is an indication that there is some co-existence. This work attempts to expand the understanding of that co-existence by combining two elements. The first, and more prominent, element, is the theoretical relationship. But the study also includes a quantitative element which allows testing and verification of both past assertions about the relationship and those assertions made in this study. By blending the theoretical and empirical at a cross-national level, the understanding of these concepts should be furthered. ' See www.unige.ch/c2d. Joel D. Sherman "A Comparative Study of Referendum Voting Behavior in Oregon, Ohio and Switzerland"(Ph.D. diss., Columbia University, 1977). The study used counties in the two states and the cantons as units of analysis. No direct link to federalism was made. Electronic correspondence with the author, Oct. 15, 1999. CHAPTER 2 - POPULISM

The task of this chapter is to take a four-step look at the phenomenon of populism.

It begins by answering what populism is, at least in the context of this study, and fiom there, moves into a discussion of the theoretical basis of populism. With that foundation built, linkages to the advocacy of direct democracy devices will be demonstrated. The chapter will conclude with a discussion of initiative and referendum.

The theoretical discussion will be broad, but will eventually narrow down to focus on two important aspects of populism. One is the concept of monism, the idea that the people are one indivisible, homogenous group. Second is the importance to populism of increasing popular participation, which leads to populist support of direct democracy.

WHAT IS POPULISM?

Building a firm definition of populism is a difficult task. A London School of

Economics conference in 1967 was devoted to finding a definition and failed to achieve a consensus.' Part of the difficulty lies in the wide array of groups, individuals and movements that have been labeled populist. Among those who have been given the title at one time or another are the Levellers of 17th-century England; the American followers of Andrew Jackson in the 1820s; the intellectual "narodniki" of 1870s Russia; the U.S.

People's Party of the 1890s; the Progressives, United Farmers movements, the Co- operative Commonwealth Federation and Social Credit in Western Canada; Lang Labor in the Australian state of New South Wales during the 1920s and 1930s; McCarthyites; 14 Peronistas in Argentina; various student movements in the late 1960s; Jimmy Carter; and in the 1990s, the Reform Party of Canada and the United We Stand America/Reform

Party movement built by Ross Perot in the U.S. Such a list contains groups of both the left and the right; of both rural and urban emphasis; of democracy, demagoguery and dictatorship. Building a meaningful and defensible definition that fits all of these groups is a near-impossible task. Canovan (1981 : 301) sums up the search for an answer best:

...if the notion of "populism'did not exist, no social scientist would deliberately invent it; the term is far too ambiguous for that. It would be far preferable to invent different words to describe the different phenomena included within it.

Two discussions that search for an answer to the question "What is populism?" are most helpful here. The first is Peter Wiles' "A Syndrome, Not a Doctrine: Some Elementary

Theses on Populism" (1969), which comes fiom that aforementioned 1967 conference.

Wiles"work remains, three decades later, as solid a description of populism as has been presented. Wiles begins with a basic definition:

To me, populism is any creed or movement based on the following major premise: virtue resides in the simple people, who are the overwhelming majority, and in their collective traditions (1 66).

From this definition, Wiles builds a list of 24 things (his term) that follow fiom it.

Various populist movements emphasize different parts of the 24 characteristics while downplaying others. Even after this exhaustive list, he still raises a handful of exceptions.

Wiles also fails to make any mention of direct democracy. But the list (167-69) is helpful and of the 24 items, the following are of particular importance here:2 15 Populism is moralistic rather than programmatic ...the actual measures asked for may

vary greatly.

Its ideology is loose, and attempts to define it exactly arouse derision and hostility.

Populism is anti-intellectual. Even its intellectuals try to be anti-intellectual.

Populism is strongly opposed to the Establishment, and to any counter-elite as well. It

arises precisely when a large group, becoming self-conscious, feels alienated fiom the

centres of power.

...p opulism avoids class war in the Marxist sense. Though certainly class-conscious, it

is basically conciliatory and hopes to convert the Establishment.

Populism can be urban...The (British) Labour Party (note that it is not called socialist)

counted among its origins trends that can only be called urban populism.

Populism opposes social and economic inequality produced by the institutions it does

not like. But it accepts the traditional inequalities due to the way of life of its own

constituency.

These "things" will show themselves at various points in this chapter.

Nonetheless, it is prudent to narrow this list even further, and to emerge with a few key points. First, Wiles states in his definition that to a populist, "virtue resides in the simple people, who are the overwhelming majority." The people are seen in broad terms. Wiles indicates that class divisions are avoided, an idea also put forward by Brugger and

Jaensch (1985:8), who write that populists believe, "the state has no inherent class nature and is merely the instrument to be used by whichever group or class is dominant." In the place of class analysis, the people are seen as held down by elites, leading to the anti- 16 intellectual, anti-Establishment nature of populism. Canovan (1 981:294) sums this up

neatly, agreeing with Wiles that populism is a syndrome of varying elements, but finding

two "universally present" elements - an "exaltation of and appeal to 'the people"' and anti-

elitism.

Despite her rather mocking conclusion cited in the earlier quotation, Canovan

provides the other discussion on defining populism, building a typology that can be used

to narrow down what sort of populism is being examined in this study. Canovan (1981:7)

claims "we cannot hope to reduce all cases of populism to a single definition." She goes

on to list seven "types" of populism aggregated within two "families" (1 98 1: 13). Three

of the types are listed under "agrarian populism" - fmers' radicalism, peasant

movements and intellectual agrarian socialism. The other four are "political populism" -

populist dictatorship, populist democracy, reactionary populism, and politicians'

populism.

Canovan admits there is overlap between the two broad families, and even

between specific types. For example, the fmers' radicalism that is best exemplified by

the Populist Party of the U.S. in the 1890s contained elements of populist democracy in

its call for referendums and participation, but also of politicans' populism, which

Canovan (198 1: 13) describes as "broad, nonideological coalition-building that draws on the unificatory appeal of 'the people'." The populism this study examines does not cover all seven types. It focuses on the use of referendums and initiatives in four democratic, industrialized countries. Thus we can quickly drop peasant movements, the intellectual agrarian socialism of some 19th century Russians, and populist dictatorships such as Juan A

Peron's Argentina With somewhat more consideration, we can also drop agrarian

radicalism and reactionary populism. Agrarian radicalism may have provided some of

the early foundations of what will be considered here and may be mentioned fiom time to

time, but it is not at the core of the study, as the range of issues considered here shall be

of a broader scope than simply agrarian concerns. As for reactionary populism, one

might argue that some populist groups of the present (such as the Reform Party of Canada

or the Zurich branch of the Swiss People's Party) fit this mold. But this seems to be a

labeling that is devoid of any real content germane to the subject of this study; such

groups also fit in within other categories. Thus we are left with two types - populist

democracy and politicians' populism.

The concept of the people embodied in what Canovan calls "politicians'

populism" is critical. Seeing the people as an undifferentiated whole provides us with the

theoretical tension with federalism described in the introduction. The label "politicians'

populism" may not be entirely accurate, however. Granting the act of building a broad,

non-ideological coalition solely to those already within the political system is too nanow.

Ross Perot made such an attempt fiom outside the usual U.S. political system in 1992.~

Calling such action "people's populism" makes the label redundant. Whichever label one

applies, this sort of populism is the kind most evident today in modem democracies, and

is at the core of the populism being examined here.

Once the people are seen as an undifferentiated whole whose will is law, there are two ways of operationalizing that will. One is through authoritarian leadership in the name of the people, as with Juan Peron. The other is through populist democracy. 18 Through direct democracy devices such as the referendum and initiative, the people are

given decision-making power, and thus, the direct ability to make their will law. It is the

latter option that is the object of this study.

Through Wiles providing some characteristics of what populism is and the use of

Canovan's typology, we have established the boundaries of what shall be studied in this

chapter. It is conceded that the defkition of populism has been narrowed. Nonetheless,

the narrowed definition is consistent with the sort of populism witnessed in developed,

democratic societies at the end of the 20th century. Populism in the present context, as it

shall be defined here, seeks governance consistent not with any particular ideology but

rather with "the common sense of the common people," which is best found through the

use of direct democracy. The task at hand is to present the theoretical underpinnings of

what lies inside those boundaries.

Rousseau and the "General Will"

Populism is not rich with deep theoretical works. As Wiles implies (1969: 167),

the idea of formally articulated theory may smack of elitism to a populist, who would

rather see discussion brought down to a more "practical" level. This would be particularly

true in earlier times, when the chasm between educated elites and uneducated masses was

wider than it is today.4

Even so, one philosopher, Jean-Jacques Rousseau, can be said to have been the

first to present some of the principles upon which populism is based. Though Rousseau's \ writings are varied and it would be stretching to actually call Rousseau a populist, he 19 does present some basis for what is now called populism. In Discourse on the Origin of

Inequality, Rousseau argues for a more egalitarian society. He longs to be able to reverse

"progress" and its incumbent division of labor. According to Canovan (1981:241),

Rousseau argues that "man is best when he is closest to nature in societies that are simple,

unrefined and egalitarian," and that people in those societies "were also wiser than the

sophisticated individuals who despised them." The idealization of the common people,

and to a lesser extent, the rejection of progress, are central to the populist ideology.'

Wiles' definition is consistent with this idealization.

The idea that society should be seen as an undifferentiated whole also can be

found in Rousseau's writings. When society is "whole", it is much easier to find the

general will Rousseau discusses in the Social Contract. It is not a huge leap to see the

"smaller societies" in the following quotation as elites:

...when private interests begin to take the lead, and smaller societies have an influence on the greater, the common interest changes and finds many opposers: there is no longer unanimity of opinion; the general will is no longer the will of all; everythmg is contested; and the best advice is never adopted without much dispute and opposition (1 762 [1947]: 93).

As will be explained later in this chapter, Rousseau viewed intermediary institutions as useless. The will of the people cannot be represented; laws must be made directly by the people. Electing representatives is roughly equated to making oneself a slave to those representatives because of the obligation to follow the laws they create. Voting is a responsibility of citizenship, thus giving up voting to a representative is to give up citizenship. Much of Rousseau's writings are contradictory; at some points he expressed great faith in the common people, in others he calls for the introduction of a lawgiver to

preside over them (Canovan, 1981 : 243). Further, while Rousseau clearly has much in

common with later populist thinkers, there really is not a lineage linking the two. Rare is

the reference by a populist of the last 100 years to Rousseau. There may be agreement,

but that agreement appears to have come independently, or otherwise without due credit.

Nonetheless, Rousseau's glorification of the people as a simple, undifferentiated whole is at the core of current populist thought.

The virtue of the "common people"

Populists believe not only that power should lie with the people, but that the common people are virtuous. Wiles' definition states this without reservation. Edward

Shils, a critic of populism, argued that populists see the people as not only the equal of the rulers, but in fact better (cited in Allcock, 1971). Nowhere is the belief in the common people presented more clearly than in one of the principles of the Reform Party of Canada:

We believe in the common sense of the common people, their right to be consulted on public policy matters before decisions are made, their right to choose and recall their own representatives and to govern themselves through truly representative and responsible institutions, and their right to directly initiate legislation for which substantial public support is demonstrated (Reform Blue Book, 1991).

Reform was hardly the first populist movement to make such a statement, in fact it does so in almost polite, passive terns compared to past movements. The Jacksonian

Democrats of the 1820s "asserted the rights of ordinary citizens against the 'gentlemen"' (Canovan, 1981 : 176). Turn-of-the-century populists rallied around "The Great

Commoner," William Jennings Bryan, to whom "the common man was suffering,

meritorious and heroic" (Koenig, 1971: 456). Australia provides the example of Jack

Lang. In reference to one of candidates of the New South Wales Labor Party in 1931 it

was said, "For (Eddie Ward) the world was divided into 'baddies', the Money Men who

were generally liars, fornicators and drunkards, and the 'goodies', like Lang, who wanted

to relieve the sufferings of the people" (Clark, 1987: 367).

One group which provided a detailed critique of the gap between the people and

"others" was the United Farmers of Alberta. In UFA writings, the people were glorified,

particularly fanners, whose "life-style and moral vantage point contrasted with the urban

tendency to distract people fiom honest pleasures" (Laycock, 1990: 74). Further, the

people's goodness was to be seen in contrast to the evil of a series of other groups and

influences. Thus, to a UFA theorist, the world was people versus politicians, who were a

breed apart from the people; people versus political parties, which divide society into

waning groups; people versus autocracy, "a general term embodying all anti-democratic

practices practiced by established political and economic elites;" people versus

plutocracy, the general unfairness of the prevailing political economy; and people (and

God) versus Mammonism (Laycock, 1990: 74-79).

It should be noted here that some political scientists have taken a view quite opposite to that of the populists. "Elite" theorists of democracy, whose influence was at its peak during the 1950s and early 1960s, argued that educated elites make the best decisions for society, and that the people, rather than being virtuous, are anmgbut. Democracy, then, should be limited to the periodic selection of leadership. Participation was good, but only if not taken too far. Almond and Verba (1965) argued in the classic work, The Civic Culture, that the most stable democracies are those which, despite their participatory elements, retain a degree of deference and apathy.

Much of this school of thought emerged in the shadow of the Joe McCarthy-led

"witchhunts" of the early 1950s. Commentators like Hofstader, Shils and Lipset then made linkages with the supporters of McCarthy and the turn-of-the-century adherents of populism. McCarthy was thus labeled as a case of "populist extremism," and populism itself was discredited (Lipset, 198 1: 169-73). Lipset provides the starkest rejection of the virtue of the common people in Political Man, questioning the ability of those of low socio-economic status to accept democratic norms, saying:

Acceptance of the norms of democracy requires a high level of sophistication and ego security. The less sophisticated and stable an individual, the more likely he is to favor a simplified view of politics, to fail to understand the rationale underlying tolerance of those with whom he disagrees, and to find difficulty in grasping or tolerating a gradualist image of political change (198 1: 109).

If the populist view of the virtue and "common sense of the common people" provides us with one end of a spectrum, Lipset's view starkly provides us with the other, as he argues that, compared to more-educated elites, the working classes are more intolerant and are more pre-disposed to authoritarianism than to denlocracy.

Another argument made against populism is mathematical. In Liberalism Against

Populism, William Riker used social choice theory to argue that populist theories were unworkable. Riker said populists believe that "participation in rule making is necessary 23 for liberty" (1982: 12), thus, "what the people ...want ought to be social policy," and "the people are fiee when their wishes are law" (1982: 238). This is an accurate assessment of populist theory. Riker, however, using a series of mathematical models and cases drawn from elections, says social choice theory tells us that we cannot determine what the wishes of the people are. Voting does not determine a "common will," it merely tells us which of the presented alternatives have been chosen. If a different method of voting had been used, it is possible that a different outcome would have emerged fiom the same set of options. Because populist theory depends on voting to tell us what the will of the people is, populist theory is flawed. Riker then concludes that the periodic choosing of leaders is as much as democracy can be expected to do (1 982: 9-1 1,241-244).

Distrust of the people dates back even to the American founding. In Federalist 5 1,

Madison raised concerns about "the tyranny of the majority" in arguing for the separation of powers in the U.S. Constitution. In fact, despite the populist rhetoric of "We the

People" in the preamble to the Constitution, the U.S. system was originally designed to keep the people fkom having too great a direct say. The House of Representatives was directly elected; but Senators were chosen by state legislatures and given fairly long (six- year) terms; the President was selected not by the people, but rather by an Electoral

College; and the Supreme Court was placed entirely outside the reach of the people.

"Elite" theorists and others who feared the "tyranny of the majority" saw representative institutions as a way of protecting decisions fkom the deficiencies of the people. It is not surprising that populists, who saw the people as virtuous, often saw those institutions in a more negative light. This tension between populism and representative democracy was a central element in populism's support for direct

democracy.

The critique of representative democracy

The populist call for direct democracy arises out of populism's criticisms of

representative democracy. These criticisms are based on a feeling that "'special interests'

...tend to dominate the political process to the detriment of 'the people"' (Canovan, 1981:

178). At its most basic level, representative democracy indicates that there is a separation

between the electorate and the decision-making process. The electorate does not make

decisions itself; instead it chooses representatives who will make those decisions. The

linkage between the representative and the electorate, or in populist terms, "the people," lies in periodic elections, in which the representative must return to the electorate seeking to serve another term of office. Populist demands for more direct democracy arise when some group believes the linkage has weakened to the point where the representative is no longer listening to the electorate, but rather to lobbyists, special interests, or party leadership (Cronin, 1989: 10). To a hard-line populist, "politician" is almost an epithet. In describing the views of early Canadian populists, David Laycock says they believed

"politicians were a breed apart fiom 'the people1, intent on sustaining social-political impotence for honest producers" (1990: 75). This view was particularly strong in the agrarian populism movements in the West during the first half of this century. Those groups were seeking greater protection against powerful special interests, but those same 25 interests, namely railroads, corporations and trusts, were seen as controlling the political

parties and thus the legislatures (Cronin, 1989: 44-6).

Agrarian populism tended to be based in the economic distress of the farmers, but

the call for more direct democracy has not been limited to such groups. A change fiom

the politics of "power" to the politics of "participation" was a key call of the student

movements of the late 1960s (Canovan, 1981 : 186). It is a part of the appeal of the "New

Social Movements," such as environmentalism and feminism, although such movements

have tended to stress participation through activism, rather than calling for direct

democracy. The Reform Party has also argued that the current representative system has

allowed for policies that go against what the "people" want. In Canada, the Goods and

Services Tax instituted by the Progressive Conservative government in 1991 served as a

lightning rod for such discontent.

Populists, then, see a division between the elite-driven outcomes of legislatures

and the views and desires of the "common" people. Despite the differences in desired

outcomes among various populist groups, the perceived root cause of those desires not

being met is consistent - the elites, and more specifically, the elected representatives, are

not responding to the people. To return to Rousseau for a moment, this is quite in line with his view that representation was illegitimate, that only the people themselves could make law:

Sovereignty cannot be represented for the same reason that it cannot be alienated; its essence is the general will, and that will must speak for itself, or it does not exist: there is no intermediate possibility. The deputies of the people, therefore, are not and cannot be their representatives; they can only be their commissioners, and as such are not qualified to conclude anything definitively. No act of theirs can be a law, unless it has been ratified by the people in person; and without that ratification nothing is a law (1762 [ 19471: 85).6

Also tied to the critique of the failure of representative democracy is the populists'

distrust of parties. There is a two-fold reason for this - and both folds deal with the

inconsistency of political parties with the concept of society as an undifferentiated whole.

As stated by UFA theorists, parties seek to divide. Most party systems are based, at least

in part, on socio-economic class, although some have their roots in region, ethnicity or

religion (Lipset & Rokkan, 1967). Analysis that finds this division counter-productive

has not been limited to populists. George Washington exhorted the U.S. to avoid the

problem of "faction" in his Farewell Address in 1797. Some Afiican states have

attempted to overcome ethnic division through one-party systems. Julius Nyerere, the

first leader of Tanzania, once wrote, "Where there is one party, and that party is identified

with the nation as a whole, the foundations of democracy are herthan they can ever be

where you have two or more parties'' (Cliffe, 1967: 14-15).

Even a one-party state, however, falls prey to the second fold of the populist

critique - the separation of elite leadership &om the masses. This phenomenon was best

identified by Roberto Michels, who developed the "Iron Law of Oligarchy." Michels

(1962) says for any party or other group to operate efficiently, it must have organization.

Within that organization, leadership emerges. This leadership may be in touch with the masses when it is chosen, but soon distances itself and develops a different set of interests. The leadership maintains its power through access to greater information and through psychological pressure, such as threatening to resign, and the organization becomes an oligarchy, led by those at the top in their interests, rather than in the interests

of the whole.

This sort of analysis is common in populist thought. Argersinger (1974: 308) cites

American populist Percy Daniels as saying in an 1898 speech,

Parties as they exist today are bellowing imposters and organized fiauds, sowing little but deception and garnering little but spoils and comption.... They are either reliable machines of plutocracy and the corporations, or they are the handy tools of hypocrites and harlequins....

Even so, most populist movements are forced by the nature of the political system to become formal parties. This is particularly true of parliamentary systems, but also is the case in the U.S. Congressional system, in part recently because of the advantage given to parties in campaign financing. The choice of operating effectively or maintaining principles can split a new party. The more common response is to adapt to the "rules of the game" hoping to change them at some later point.

One example of the hard-line response, of refbsing to yield to the logic of the system, is the Progressive Party in Canada. Following the' 192 1 general election, in which they won 64 seats, the Progressives were in numerical position to become Canada's

Official Opposition, but declined, as part of a platform that also resisted party discipline and formal leadership. Within a few years, the party disintegrated, with many of its MPs becoming Liberals, and by 1930, the remnants were picked up by newer protest parties.'

The anti-elitism of populism makes its adherents skeptical of representative democracy. Borrowing from Michels, elected officials are seen as elites who carry out only their own agendas, not the agenda of the people. Efforts to operate without

leadership in the current system, such as the Progressive Party, have proven to be

ineffectual. How then can the will of the people be brought into decision-making? The

answer for populists lies in going around representative democracy. For democratic

populists, it lies specifically with direct democracy.

Populist support of direct democracy

If one accepts, as the populists do, that the "grass roots" or common people are

virtuous and how best how to govern themselves, then the next step is to provide ways

for those people to have a greater say in governance. Populists seek to change institutions

by giving a role to the people that goes beyond the mere electing of representatives and

allows them to have a more "direct" role in decision making. This is seen as a more

legitimate way of making decisions, and "all political decisions should be as legitimate as

possible" (Butler & Ranney, 1978: 24).

Movements of the type Canovan labels populist democracy seek to reform

democratic institutions through the application of direct democracy devices. The three

central devices of direct democracy are initiative, referendum and recall. An initiative is legislation proposed by the electorate through petition. If enough petition signatures are obtained, the measure is put to a popular vote and, if backed by a majority, it becomes law without the need for legislative ratification or consent. Referendum refers to items sent to the people by the government for approval.' Recall is the procedure for removing public officials through a vote generated by petition (Cronin, 1989: 2). Initiative and 29 referendum, which have a direct impact on policy formation, are the focus of this study.

The study has set aside recall, as it only indirectly impacts policy and does not have the prima facie conflict with federalism that the other two devices do. More recently, the suggested devices have expanded. The "electronic town hall" concept floated by Ross

Perot is one.' In the US.,where the incumbency rate of Congressmen and other legislators has been quite high, the term-limits movement has gained momentum. The idea behind tern limits is to bring legislatures closer to the citizens by eliminating the phenomenon of the "career politician" who never had a "real job" and is thus out of touch with the electorate.1° Calls for campaign financing refom, particularly through the restriction of Political Action Committees, have also received attention. In Canada, populists have sought a change in Parliamentary rules that would diminish party discipline and allow more freedom for the individual MP. This list is presented only to demonstrate the range of ideas put forward by populists and others. As with recall, these measures will not be discussed fiuther in this study.

Cronin (1989: 10-1 1) provides six claims that populists make in favor of more direct democracy. That list, although not exhaustive, is provided here.

Initiatives promote government responsibility and accountability. If people are

ignored, they still have a means of making laws.

Initiatives are fieer from special-interest domination than legislatures, thus

providing a safeguard when legislators are corrupt, irresponsible, or dominated by

special interests. 30 Initiative and referendum provide open educational debate on critical issues that

might otherwise be overlooked.

Referendum, initiative and recall are non-violent means of participation which

fulfill the right to petition government for redress of grievances.

Direct democracy increases voter interest and turnout.

Initiatives are needed because legislators often evade tough issues and take a zero-

risk mentality.

Some of the claims are straightforward. It is very difficult to dispute that these measures are a non-violent means of participation. While one might question how "educational" it is, that these measures bring out debate on otherwise-overlooked issues is valid. There is even a thread of thinking that argues that participation itself is educational (Barber, 1984:

235).

Cronin's first and last points can be examined in tandem. In a Parliamentary system that usually produces majority govenunents, government accountability and responsibility 'already exist. Accountability and responsibility are less clear in the Swiss system, even less so in U.S. As Fiorina (1 980: 44) has written, "In contemporary

America officials do not govern, they merely posture." The initiative does give the electorate a means of taking matters into its own hands and doing, as has been stated earlier, an end nu around the legislature. However, the usefulness of the initiative in this context may be cancelled out by the referendum, as referendums can be used to "pass the buck" on tough issues (Cronin, 1989: 185). Pierre Trudeau once mused about such an 3 1 idea, saying "it might be time for the Government to throw a few hot potatoes back (to

the people)" (Boyer, 1992: 43).

Point five claims direct democracy increases voter interest and turnout. The

evidence of this is weak. In Switzerland, voter turnout in referendums has fallen below

50 percent, perhaps because of "voter fatigue" fiom holding the referendums every three

months (Canovan, 198 1 : 2 1 1). The American model places the issues on the ballot at the

same time as elective offices. Only on very hotly contested issues does turnout increase

from its normal levels (Cronin, 1992: 67-68). , Cronin's most important point is that initiatives are freer fiom special interests.

This is a central point because one goal of the populists is to return power to the

"common" people. But in many U.S. states, particularly California, the costs of running

an initiative campaign are high enough that those best positioned to do so are the same

groups already lobbying the legislature, the groups that the initiative process is supposed

to thwart (Canovan, 198 1 : 2 16). There is also the issue of campaign spending. While

money shows little influence on getting initiatives passed, it has shown an ability to

successfblly oppose initiatives (Cronin, 1992: 109-1 10). Nonetheless, if well-placed interests have been able to use initiatives, so have smaller ones.

The drive for direct democracy, and the analysis of elites being out of touch with the people that is behind it, does not emerge as some sort of altruistic, value-neutral idea.

There is an assumption among its advocates that changing the system will change the outputs to the benefit of those seeking the systematic change. Agrarian populists, for example, felt they were being short-changed by a system catering to urban-based 32 economic elites. Cronin quotes an Oregon populist fiom the turn of the century as saying,

"The important thing was to restore the lawmaking power where it belongs - in the hands

of the people. Once (you) give us that, we could get anything we wanted" (1989: 49)."

Patrick Boyer's point that in Canada the people have been content to see the system as

something far fkom themselves, that Canada has been a "timid democracy," supports this (1992b33). If outputs. are satisfactory, there appears to be little discontent with elite leadership. Canadians may feel far fiom their government, but if they are happy with the

outputs that govemments provide, they are unlikely to seek change. It is only when some

segment of the population sees itself as unrepresented and disadvantaged that populism's

call for direct democracy really takes hold. In the United States and Australia, the

populist impetus that led to the introduction of direct democracy emerged as a reaction to

the worldwide economic downtum of the early 1890s. This downturn hit the hardest on

fanners, and it was among the farmers populism was strongest. The case in Switzerland

is somewhat different, but it is worth noting that expansions of direct democracy in 1848,

1871 and 1891were all as the result of a group (Radicals in 1848, Catholics in 1871 and

1891) which had seen itself as being outside the decision-making process taking advantage of an opening to change the system and gain a foothold within it.

The use of direct democracy

Many democratic states have quite limited experience in using direct democracy.

In most European states (Switzerland being the crucial exception), the only direct democracy arises in very occasional referendums. Membership or other issues involving the European Union have been the most common topics of referendums in ~urope,'~ although constitutional issues have also been raised, most notably in France, Ireland and

Spain. Only a handhl of referendums have been held in New Zealand, but an important referendum in 1993 changed the electoral system fiom a single-member, first-past-the- post system to a mixed system of single-member districts and proportional representation along the lines of Germany.

Despite only occasional use, referendums have had a dramatic impact on Canada.

The referendum on the Charlottetown Accord in 1992 is the most recent nationwide example. The people were asked to vote on a large set of amendments to the Constitution.

While exactly what level of support was needed was unclear, it was argued by many that the referendum would need approval in all 10 provinces. In the event, the package of amendments was rejected by a 55-45 margin, and majorities of voters in only four provinces supported it. The referendum was the third national referendum in Canadian history. The only other national referendums were on prohibition of alcohol in 1898 and on military conscription for World War I1 in 1942. It is worth noting that both passed nationally but were massively rejected in Quebec, which points up one of the difficulties of national referendums in Canada - a matter of enough importance for a referendum is often one the "two solitudes" see differently.13 A few provincial referendums have also been central to Canadian political history - the two-stage referendum in 1948 which brought Newfoundland into Confederation and the sovereignty referendums in Quebec in

1980 and 1995 - and every province has had at least one referendum on prohibition

(Jackson & Jackson, 1992: 489). 34 Switzerland has used direct democracy more extensively than any other country.

The Swiss vote on national referendums and initiatives every three months and also have many referendums and initiatives at the cantonal level. It can be argued that direct democracy emerged out of the nature and history of the Swiss Confederation, rather than some populist urge. This has not prevented populists fiom praising the Swiss model, and we can identify at least two countries -Australia and the United States - where populist pressures led to the adoption of direct democracy devices.I4 Furthennore, Steiner, perhaps the preeminent North American-based scholar of Swiss politics, has provided descriptions of Switzerland that would be very consistent with the model populist state:

...we do not think we are a parliamentary democracy. We call ourselves a direct democracy; therefore constitutionally it is clear that the ultimate say is with the people. The ultimate power, not only to elect representatives, but to make decisions, is with the people (in Ranney, ed. 1981: 6).

In Australia, all proposed constitutional amendments are brought before the people in conjunction with federal elections. For approval, these referendums must be passed not only by 50 percent of the overall voting electorate, but also must achieve a majority in at least four of the six states. From 1906 to 1988,40 constitutional questions were put before the electorate. Only eight were approved. Four others received a majority of the vote, but did not pass in enough states to gain approval. Additionally, in November of

1999, Australian voters rejected referendums which would have made Australia a republic and introduced a preamble to the constitution.

In the US.,there are no federal direct democracy devices, but they are common at the state level. A referendum is required to approve constitutional amendments in 49 of 35 the 50 states - the exception being Delaware. Some 26 states, mostly in the West, use the

initiative. Many of these states adopted the initiative during the Populist surge of

influence around the tum of the century (Cronin, 1989: 5 1). Similar populist pressures led

to U.S. Senators, who had originally been chosen by state legislatures, being directly

elected in each state through the ratification of the 17th Amendment in 1913.

The most extensive use of the initiative has been in California. Californians voted

on 264 initiatives in the 1980s (Boyer, 1991:144). Perhaps the most famous initiative

occurred in 1978, when Proposition 13 slashed property taxes in half and triggered a

spree of tax-cutting initiatives across the U.S (Cronin, 1989: 3). In 1993, voters in

Washington state rejected an initiative that would have rolled back taxes, but passed

another that put a cap on increases in state revenues. Some issues are more esoteric. In

one of the more novel initiatives, a Nevada county legalized brothels in 1975 (Barber,

1981 : 283-4). Other initiatives deal with institutional reform. A number of states passed

term-limit initiatives in the early 1990s. Fourteen did so in 1992. In 1994, six states

passed term-limit initiatives and only in Utah did a term-limit proposal fail.

Initiatives are generated by petition drives. In the U.S. model, the petition, which

has the exact wording of the proposed law, must receive a designated percentage of the

total vote in,the last election of a state governor. That percentage is typically eight percent, although in some states it is five or 10 (Cronin, 1989:62; Boyer, 1992a: 145). If suflicient petition signatures are gathered, the measure is placed on the ballot in the next statewide election. In Switzerland, the process is somewhat different, with a fixed number of signatures needed (1 00,000 at the federal level) and with the measure coming to ballot 3 6 separately &om legislative elections (Canovan, 1981 : 200). In an alternative form of the initiative used in some U.S. states, the "legislative initiative," or indirect initiative, the petition goes to the legislature, which can pass it immediately, or send it back to the people, either on its own or with a counter-proposal.

Canada has virtually no experience with the initiative, but the process has recently gained a toehold in two provinces. In 1991, Saskatchewan passed the Referendum and

Plebiscite Act; initiatives may be brought to the electorate if 15 percent of the voters sign a petition, but the results of that vote are non-binding (Boyer, 1992a; 139). Also in 1991,

British Columbia voters gave 81 percent support to the idea of initiatives. A 1993 legislative committee recommended some very stringent requirements, beginning with a

10 percent threshold on signatures that would need to be achieved in all 75 provincial constituencies. If the initiative reached the voting stage, it would need a double majority - a majority of all eligible B.C. voters, not just those actually voting, and a majority in two- thirds of the provincial constituencies (Matas, 1993).

The Reform Party advocated a national initiative system for Canada, calling for a rather low petition requirement of three percent of the electorate or, according to 1991 figures, about 780,000 signatures (Manning, 1992: 325). The party claimed that "(w)ith the initiative in place, citizens can get around obstructive politicians and strategically located special interests to ensure that matters relating to the general interest are placed before the people." Reform would also seek constitutional changes to make such initiatives binding law-making procedures ("Reform Party Caucus Issue Statement 36," July 16, 1992). The Canadian Alliance has retained the call for initiatives in its platform,

although not with a specific threshold percentage of signatures.

A hybrid of the initiative and referendum is the facilitative referendum, sometimes

called the popular or direct referendum. In a facilitative referendum, signatures are

gathered calling for the repeal of recently passed legislation. If sufficient signatures are gathered within a prescribed length of he,typically 90 days, the legislation is sent to the electorate for its acceptance or rejection. The device is used in Switzerland at both the federal level and in the cantons (Delley & Auer, 1986: 86-91) and in 23 U.S.states (Book of the States, 1996-97: Table 5.20), but has no presence in Canada nor Australia.

Differences in the populist nature of referendums and initiatives

While referendums and initiatives both find support in populist thought, it can be argued that some types of ballot measures are more populist than others. This is best determined by the level of popular involvement in placing a measure on the ballot.

Broadly speaking, measures sent to the people at the discretion of the legislature have less popular influence than those put on the ballot by the electorate itself. The less influence and involvement the intermediary institution (i.e., the legislature) has, and the greater the direct influence and involvement of the people, the more populist the process is.

The least populist of all ballot measures are referendums brought forward at the discretion of the government. Most European referendums outside Switzerland and Italy are of this nature. While such measures are less LLpopular"than other ballot measures, they nonetheless provide greater popular participation than decisions made without 38 reference to the electorate. This is true even in cases where the vote is non-binding, as

governments have typically been unlikely to go against the popular decision. For

example, the Labour Party in Great Britain put devolution proposals to the electorates in

Scotland and Wales in 1979, but had those proposals rejected (Norton, 1994: 229).

However, in September of 1997, new devolution proposals were approved, and in light of those votes, the Scottish Parliament and the Welsh Assembly were created by statutes passed by Westminster in 1998.

Slightly merdown the populist path are referendums which the government is compelled to bring forward. For the most part, these are.referendums which seek ratification of constitutional changes, such as in Australia. These remain under government control, however. It is the government which decides to put the issue itself forward, but rather than having a choice about going to the electorate, it is compelled to.

Somewhere in the middle are facilitative referendums. The impetus here remains with the legislature, but is the people, through petition drives, who determine whether or not the matter is brought to a popular vote.

The initiative is the most populist of the direct legislation devices, but there is some differentiation of levels of populism in various types of initiatives. The indirect

initiative, used in Washington state and Zurich, gives the legislature the opportunity to 8 approve the legislation instead of sending it to the full electorate. In this way, it maintains some involvement of the legislature. The direct initiative, with little or no legislative involvement, is the most populist of the processes. The heaviest populist influence, then, would be present where the direct initiative is present. CONCLUSION

The task of this chapter was to take a four-step look at populism. Through the use

of Wiles' "syndrome" discussion and Canovan's typology, it was established that

populism is a strain of political thought which glorifies the common people, and thus has

negative attitudes towards elites and elite leadership. Out of this, two main themes of

populist thought emerge. First, populism is monist, thus it sees the people as an

undifferentiated whole. Second, the particular brand of populism under study here, which

is the most common in democratic countries, claims the people can best express their will

through the use of direct democracy. This emerges fiom a rejection of intermediary

institutions controlled by elites, as elites are viewed as out of touch with the "common

people" and therefore use their position as elected officials to further their own interests

ahead of those of the people. The solution recommended by populists is the use of direct

democracy devices such as referendum and initiative, which place decision-making power in the people. With the appeal of direct democracy to populists established, the actual usage of referendums and initiatives was discussed.

Populism's view of the people as an undifferentiated whole and its celebration of the grass roots stands in variance with the topic of the next chapter - federalism.

Federalism recognizes, enshrines and perhaps even celebrates the differences of constituent units of the state. Federalism is often in place because of an explicit understanding that an undifferentiated whole does not exist. Furthermore, federal systems are characterized by negotiations and agreements between governments. Because of this, the process of federal governance can become highly technical, the

purview of the bureaucratic and political elites so despised by populists.

- 'The conference produced Ionescu, Ghita & Ernest Gellner, eds., Populism: Its Meanings and National Charcteristics (London: Weidenfield & Nicolson, 1969).

The bulleted items are quoted directly from Wiles, but the entire list of 24 is not provided here.

It can be argued that the act of such coalition building transforms the builder into a politician.

One exception to this general rule regarding formal articulation has been Preston Manning.

The rejection of progress and its implied call for a return to a simpler time is not central to the argument here, but can be seen on both the right and the left. For example, concerns about the increasing diversity of the population or the globalization of the economy both fit this rubric.

Rousseau is not, however, suggesting a majority vote along the lines of the ballot measures examined here. Rousseau argued the general will would emerge, something along the lines of a unanimous consensus. Further, Rousseau did not see "public opinion" as emerging from a calculus of individual desires as Bentham might (Price, 1992: 10- 15).

'Echoes of such debates can be found within the Reform Party and the 2000 decision to transform itself into the Canadian Alliance. Reform, however, was a fonnal party fiom its inception.

Some writers distinguish between referendums and plebiscites by saying that the former are binding and the latter are merely advisory (See Boyer, 1992b:23-25). The author carries some sympathy with this view. However, because the vast majority of cases in this study are binding (the primary exception being most of the Canadian cases), I shall, for simplicity sake, use referendum in both binding and advisory contexts.

Benjamin R. Barber, Strong Democracy: Participatory Politicsfor a New Age (Berkeley: University of California, 1984) 273-78, provides a detailed description of how such a process would operate.

'O Term limits can be seen as a populist device which attempts to take office away fiom "professional, career" politicians and puts those offices back in the control of "citizen legislators." George Will has discussed this at length in Restoration: Congress, Term Limits and the Restoration of Deliberative Democracy (New York: Macmillan, 1993). Conversely, they might also be argued to be anti-populist by limiting the choice of candidates available to the electorate.

" William VRen, quoted in Cronin, 49. Originally quoted in Scott W. Reed, "W.S. U'Ren and the Oregon System" (Senior honors thesis, Princeton University, 1950) 18.

"Seven of the 15 current members of the EU (Great Britain, Denmark, Ireland, France, Austria, Sweden and Finland) have had referendums concerning EU membership or some aspect of EU policy. Five countries held referendums in 1993 on joining the EU, with voters in Sweden, Finland and Austria saying yes, and voters in Norway and Switzerland saying no. l3 Of course, there was a much broader rejection of the Charlottetown Accord. "Some evidence of this fertile ground can be found in the opening lines of the U.S. and Australian constitutions. The U.S. constitution begins in a very populist tone with the words, "We the people," even though the constitution itself attempted to put checks on "popular" rule. The Australian document at least makes a mention of the people, saying "...the people of (the six states) ...unite in one indissoluble Federal Commonwealth under the Crown..." Contrast this to the Canadian BNA Act of 1867, which makes no mention of people, only provinces. Furthemore, a JCPC ruling on a 19 16 Initiative and Referendum Act passed by the Manitoba legislature said, "In Canada there is no sovereignty in the people" (Morton, 1944: 287). CHAPTER 3 - FEDERALISM

This chapter shall follow a format similar to the previous chapter, which

introduced populism. It begins by answering the question of what federalism is, moving

from there to a discussion of its theoretical basis. Once again, the general discussion will

be broad, but will give added emphasis to facets of federalism important to this work, in

particular divided sovereignty and federalism's anti-majoritarian aspects.

WHAT IS FEDERALISM?

Discussion of the definition of federalism is not nearly as fractured and wide- ranging as the discussion of the definition of populism. Nonetheless, evolution of the use of the concept has been underway, particularly within the past two decades.

Paradoxically, this evolution appears to have both narrowed the definition of 'federalism' and broadened it.

One can come up with a baseline definition of federalism. To wit: apolitical system in which powers are constitutionally divided between a central government and sub-national units. This sort of definition would be consistent with that seen in the introductory-level political science textbooks of some of the countries in this study (see

Greenburg & Page, 1993: 72; Gibbins, 1990: 25; Jaensch, 198 1: 28). The narrowing of such a definition took place through the hiving off of the actual institutional structures flowing out of federalism into the term 'federation' (Burgess, 1993: 4-5; King, 1982).

Thus, federalism is practiced in federations rather than, as Riker (1964) described it, 43 federalism being practiced in federalisms. On the face of it, at least, this should be seen

as advancing the precision of the concept, but it does force us to detour for a moment to

define federation and redefine federalism.

Wheare (1963: 10) defined federalism as "the method of dividing powers so that

the general and regional governments are each, within a sphere, co-ordinate and

independent." Riker (1964: 5-6) stated this with more precision, writing that in a federal

system (or, to use King's term, federation) the central government has exclusive powers

which may extend into all but one area of policy, or only one area of policy. A central

government which has powers in all areas is unitary; that which has independent power in

no area is, at the most, a confederation.

King (1982: 123), however, sees a weakness in such an approach. All states, no matter how unitary, delegate some things to their regional or local authorities. It is difficult to conceive of New Zealand, widely considered to be a unitary state, making car- parking rules a national matter.' Only a handhl of microstates could be labeled as unitary. San Marino, perhaps, would fit. As an alternative, King focuses on the federal guarantee of regional units having representation in the center in building a definition of federation:

Basically we propose that any federation be regarded as an institutional arrangement, taking the form of a sovereign state, and distinguished fiom other such states solely by the fact that its central government incorporates regional units into its decision procedure on some constitutionally entrenched basis (King, 1982: 77). 44 One might argue almost instantly that Canada falls short of this definition of federation.

The weakness of the appointed Senate may lead some to argue that there is no provincial

element to the federal decision-making power. Wheare (1964:20) concludes Canada has

a quasi-federal constitution rather than a federal one. However, this claim must be seen

to be wanting. First, the Senate still has theformal power to pass or reject legislation,

and although its members are allocated on the basis of equality of "regions," the actual

membership within those regions is still allocated on a provincial basis. Furthermore,

introduction of The Constitution Act, 1982 provided the provinces with an amending

formula in which they have a well-defined role. Provincial incorporation into the

institutions and decisions of the federal govemment is weak, to be sure, but it is not non-

existent.

The process of developing a definition for federation appears to narrow the

definition of federalism to merely the philosophy behind federation, which becomes a

catchall phrase for the use of certain types of institutional structures. But in fact, those who advanced this splitting of concepts have gone on to add a new element of confusion.

Federations, it is argued, must be based on federalism, but federalism may not always lead to federation (Burgess, 1993: 12). Such usage creates problems. Freed fiom the boundaries of formal structures that constitute federations, what are the limits of the philosophy of federalism? Frenkel(1986: 78-89) has compiled a list of some 460

"hyphenated or qualified federalisms," including "layer cake," "marblecake", and even

"bamboo fence federalism." As Jackson and Jackson (1990:241) point out, ...the concepts become so loose that any form of delegation of power can be included, with the result that the definitions become so general that federalism can mean practically anythmg to anybody.

Wheare's reservations about Canada notwithstanding, few today would question that the

four states in this study are federations. A more questionable case would be Spain, in

which a federalist philosophy has been put into practice, but without Spain necessarily

becoming a federation. Autonomy is constitutionally guaranteed to Spanish regions, but

sovereignty remains at the center. Some stretch federalism further still, claiming that

recent moves toward decentralization in France and Italy are based on federalist

philosophy, while stating clearly that those states are not federations (for example, see

Loughlin, 1986: 94). This sends us scampering back to King's point that even the most

unitary states employ some decentralization.

There is, however, a way out of this conceptual and definitional maze. Note in the

baseline definition above the mention of "constitutionally divided" powers. King and

Burgess seem willing to leave the constitutional distinction in the defition for federation

while abandoning it for federalism. The question to be asked, then, is whether federalism

is a broad philosophy, encompassing any delegation of power, or if it should be narrowed

strictly to those divisions of power that have some constitutional basis.' The answer may

lie with Elazar and his discussion of the "covenant" basis of federalism.

Elazar describes covenant as one of the three ways in which political relationships form, the others being organically and through force. A covenant provides "the 46 establishment of communities of equals on an equal basis by pacts reflecting agreement and consent" (Elazar, 1994: 13). The word "federal" itself comes fkom a Latin root meaning covenant (Elazar, 1994: 2 1):

While federalism is normally understood as having to do with political structures, in fact, the federal idea speaks principally to the character of human relationships. With its roots in the biblical idea of covenant, it understands humans as autonomous equals capable of entering into covenants to establish the rules and institutions of their self-government ....Federalism is the practical application of the covenantal way to the organization of political authority and power (1994: 5).

Covenants, then, establish the rules of relationships. These relationships are also to be carried out between equals.' A covenant is a binding agreement, similar in some respects to a contract. Duchacek (1970: 192-93)sees federalism in such terms, calling "a federal constitution ...a political compact." The effect of this may best be seen through a

Canadian example. Quebec interpretations of the BNA Act of 1867 have often seen the act as a compact, a binding agreement arrived at between consenting parties. Therefore, any change must be approved by all parties. Some say compact theory indicates the BNA

Act was a compact among all provinces (Black, 1975), others say it indicates agreement between "two founding peoples."

Like a covenant, a constitution is a binding agreement. Constitutions structure relationships between both governments and people and, in a federation, between governments and governments (Cairns, 1988). The U.S. constitution, for example, defines state-and-center relationships through the division of powers and Senate 47 representation, and defines the relationship between the government and the governed

through the Bill of Rights. At the bare minimum, a federation cannot exist without a

constitution which lays out the covenantal relationship between the national government

and the regional sub-units.4 If federalism is based on covenant, the structures built by

federalism must be binding and consensual. To argue that the French effort to devolve

power to the departments and communes is somehow federal is spurious. The

government in Paris holds all the cards. If it decided tomorrow to bring all decision-

making power back to the central authorities, it could do so without the consent of the

sub-national levels. France is practicing decentralization, it is not practicing federalism.

There is no covenantal relationship between Paris and the countryside. It is here, then,

that the definitional critique of federalism put forward by King and Burgess falls short.

Nonetheless, their analysis is useful. King's definition of constituent units having

a voice in the center is critical to an understanding of federalism, and to the analysis in

this study. It also leads to an understanding of the convenantal nature of federalism.

Thus, we can develop a more precise definition of federalism, which incorporates the

baseline definition, while addressing the concerns of King and Burgess, and also

incorporates the covenantal perspective of Elazar:

Federalism is the covenantalpractice of constitutionally dividing power between a central government and sub-national units, and granting those sub-national units a formal role in the decision process of the national government.

This two-part definition provides a usefbl framework for the analysis carried out in this work. The division of powers appears, prima facie, to be a counter to the populist perspective of society as an undifferentiated whole out of which a popular will can emerge. This apparent conflict will be discussed more thoroughly in the next chapter.

The formal role of sub-national units in the national government has a less central but still significant impact. Ideas emerging at the sub-national level, particularly those ideas which are brought to prominence through the use of direct democracy, may work their way to the larger stage of national politics. It is plausible to argue that the greater the linkages of sub-national units to the national government, the greater the probability that those ideas will move onto the larger stage.

WHY FEDERALISM?

Having established what federalism is, the next step is to ask why some states use federalism. Are there specific circumstances under which federalism emerges or is useful? If so, what are they? An alternative way of answering the "why" is to see if there are elements of federalism that can be useful for any society, or at least, any democratic society.

The circumstantial approach was the primary approach of the post-war era.

Wheare and Riker provide two of the best examples of this approach. Wheare (1964: 35-

52) asks "When Federal Government Is Appropriate." He begins by describing the baseline condition for federalism to take hold:

..the communities or states must desire to be under a single independent government for some purposes. ...They must desire at the same time to retain or to establish independent regional governments in some matters. ...To put it shortly, they must desire to be united, but not unitary (35-6).

Wheare then becomes more specific, asking when the desire for union comes about and

providing six primary factors:

The need for common military defense;

A realization that only through unity could independence from foreign powers be

secured;

Economic benefits;

Some prior association of the states involved;

Geographical proximity;

Similarity of political institutions.

Conveniently for this study, Wheare (1964:37) points out that in various measure, all six of these factors were present in the four federal states examined here. Wheare then asks what factors lead to the desire for sub-national units to retain some autonomy.

Riker varies only slightly from Wheare in stating the prerequisites of federalism.

He is more concise, stating that all modem federations arise fkom two pre-conditions

(1964:49) - "a willingness to compromise and a recognized need for military unity." He alternatively refers to the compromise element as the expansion condition (1 964: 12-1 3).

The expansion condition refers to a desire among a group of politicians to expand their power, but an unwillingness to do so by force. Expansion, then, must be achieved through a bargain. Riker says those desiring expansion "must offer concessions to the 5 0 rulers of the constituent units." The military condition refers to the desire of those who

. accept the offered bargain to gain greater military and diplomatic protection fiom an

external threat.

Riker stops short of claiming these are the only prerequisites for successful

federation formation, but does claim that these two conditions are necessary and present

in every successful case of federation since the U.S. Constitution of 1787, and that one or

both are missing in every failure. Thus, Riker follows a circumstantial approach that says

federalism is a viable option only when an external threat exists and that threat forces

smaller units to come together in an effort to possess greater forces capable of resisting

that threat.

The circumstantial approach is useful, but the question of "why" can also be

answered in an alternative way. The answer to "Why federalism?" might be that

federalism is a useful system, or to be more precise, a useful principle, almost regardless

of circumstance. Perhaps the best example of this comes fiom Federalist 10 and 51.

Clearly, the U.S. constitution was a creation of specific circumstances, and was more a

product of compromises than abstract principles. Nonetheless, Madison's defense of a

federal division of powers he was arguing against a few months before is a sound argument for federalism as a principle.

The central point of Federalist 10 is the need to control the effects of "faction."

Madison raises the possibility of a majority faction acting in a tyrannical manner and suggests that some anti-majoritarian control is needed to prevent this. In Federalist 10, 5 1 the primary method of control is through representative government, which is seen as

overcoming the problem of faction to a degree pure democracy cannot. Nevertheless,

there is also support for the federal principle. While positing the benefits of larger

political units in protecting liberty, Madison concedes that representatives elected fiom

too wide a range of electors might be out of touch with the needs of their constituents.

He argues:

The federal Constitution forms a happy combination ... the great and aggregate interests being referred to the national, the local and particular to the State legislatures (Federalist 10).

Madison expands on this in Federalist 39 while responding to critics who claim the new

constitution would be unitary rather than federal. Madison contends both unitary and

federal elements are present, and uses as an indicator of the federal elements the non-

majoritarian aspects of the ratification process.

...it is to result neither fiom the decision of a majority of the people of the Union, nor fiom that of a majority of the States. .. .Were the people regarded in this transaction as forming one nation, the will of the majority of the whole people of the United States would bind the minority, in the same manner as the majority in each State must bind the minority; . ..Each State, in ratifying the Constitution, is considered as a sovereign body independent of all the others, and only to be bound by its own voluntary act. In this relation, then, the new Constitution will, if established, be afederal and not a national constitution (Federalist 39).

Support for the federal principle finds additional discussion in Federalist 51. There,

Madison argues liberty is better protected by two distinct governments than by one all- encompassing government. This liberty is fbrther protected by the distribution and separation of power within those governments. Thus, a "double securityy'of rights is

created, as "different governments will control each other, at the same time that each will

be controlled by itself (Federalist 51)." This is part of a larger discussion of size in the

essay, with Madison arguing liberty is protected by the division of society itself into a wide range of component parts. This division protects minority interests from being overridden as they might be if the society came together though only one level of government. Instead,

society itself will be derived from and dependent on so many parts, interests and classes of citizens, that the rights of individuals, or of the minority, will be in little danger from interested combinations of the majority (Federalist 5 1).

Limiting the power of majorities is a central component of Madison's thinking.

Federalism is just one of the elements which achieve it, but this facet of federalism is a theme which has been emphasized by a wide range of commentators since. Livingston

(1 956: 3 10) writes "federalism is by its very nature anti-majoritarian," and Lijphart

(1985: 4) claims federalism is a "rejection of majoritarian democracy." Galligan

(1994:47-8) goes further, writing that federalism limits majorities because it "fragments power', and "enshrines complicated procedures and conflicting institutions within the democratic process."

Federalism can be seen as anti-majoritarian in two respects. First, it provides sub- national units representation in national institutions. Most often, this occurs through an upper house which represents sub-national units equally regardless of population. Pure majorities of the people, typically represented in the lower house, are diluted by this institution. In Australia, roughly 20 percent of the population has half (30 of 60) of the

Senate votes. In the U.S., less than 25 percent of the people live in states which collectively have half the Senate votes. A related point is made by Livingston (1956), who focuses on the various amending formulas of federal constitutions, pointing out the ways in which minorities can use a handful of sub-national units to block amendments.

Second, federalism can be used to remove some issues fiom a national forum and allow sub-national units to decide them. Through this, minorities, at least those spatially distributed in a favorable way, may choose policy outcomes the majority of the national community would reject.

The idea of federalism as an operating principle has been given greater play in the last two decades. Elazar (199525) argues that federalism, whether it manifests itself in federations or confederations (such as the European Union), has supplanted statism, and its by-product, the nation-state, as the dominant paradigm in international relations. He writes "the world as a whole is in the midst of a paradigm shift fiom a world of states .. . to a world of diminished state sovereignty and increased interstate linkages of a constitutionalized federal character." Burgess (1993:112) even raises the possibility of federalism as an ideology, although he does so in a very narrow way, saying "it is not a doctrine which trumpets universal, a priori truths." In fact, this sort of view of federalism would be unusual, indeed. However, Verney (1995:8 1-3) does provide a 54 somewhat broader view than Burgess. Vemey claims federalism should be seen as an "-

ism," little different fiom socialism or liberalism because it "is based on a normative

concept of human nature and social relations." Simeon and Swinton (1995:ll) come

closest of all to a universalist stance, stating that federalism serves as a "middle ground"

between two undesired poles of "excessive centralization and the fiagrnentation of the

political world into a multiplicity of separate, ethnically homogeneous nation states."

While both approaches to the question of "Why federalism?'are valid, viewing

federalism as a useful principle provides more utility for the analysis in this study.

Federalism is a fact in the four countries selected for examination; the circumstances of

its development are tangential to the fact of its existence. Treating federalism as a

principle, particularly in respect to its anti-majoritarian aspects, moreover, provides us

with a counterpoint to populism as a principle. It is that tension between these two

principles that is the central focus of this study.

A FEDERALIST CONCEPTION OF SOCIETY

Federalism does not see society as one single entity. The federal principle assumes fiom the beginning that multiple groups make up society. Federation provides a method of granting formal status to these groups, usually on a temtorial basis.

In fact, territory can be the only differentiation between groups. There is little difference in the lives of those in Lewiston, Idaho and those across the Snake River in

Clarkston, Washington. The two towns were settled by similar people at similar times. h 55 Their residents speak the same language, have a highly intertwined economy, share a

micro-climate unique to the area and receive news fiom the same media sources. But the

two towns and their residents have been defined as belonging to different federal groups

within the U.S. political system. Lewiston belongs to Idaho, and interacts at the state

level politically with Boise, Twin Falls, Moscow and other places. Clarkston belongs to

Washington, and interacts with Seattle, Bellingham, Yakima and other centers. Perhaps

more striking still is the Canadian prairie town of Lloydminster, divided down its main

street with Alberta on one side and Saskatchewan on the other.

Artificial as the boundaries of these communities may seem, they do develop

legitimacy over time. In an effort to justify their existence and extend their reach over

society, sub-national governments often seek ways to legitimize the seemingly artificial

communities contained within their j~risdictions.~Speaking of Canadian provincial

governments, Alan Cairns (1989: 145) pointed out "their sources of survival, renewal,

and vitality may well lie within themselves and in their capacity to mold their environment in accordanceivith their own governmental purposes." He adds "the governments of Canadian federalism have endowed the cleavages between provinces, and between provinces and nation that attended their birth, with an ever more comprehensive political meaning."

As the impact of government on day-to-day life has increased, the importance of sub-national divisions has increased. People in Clarkston look to Olympia for services, and in the same way, those in Lewiston look to Boise. The greater the level of services 56 provided, the greater the level of importance that linkage to the capital holds. Over a long period of time, the legitimacy of these links grows and is even taken for granted. Even non-governmental, or to extend this a step further, non-political organizations may reflect the boundaries created by federalism. As Donald Smiley (1987:4) explained,

...Canadians organize themselves as such not only for the purposes of government but as Presbyterians, manufacturers, university teachers, playwrights and so on. Similarly, there is not only an Alberta political jurisdiction with its territorial boundaries and legislative powers protected by the Constitution but also an Alberta Red Cross, an Alberta Teachers Association, an Alberta Conference of the United Church of Canada, an Alberta Chamber of Commerce and an Alberta Federation of Labour.

The territorial division of sub-national units may be linked to the spatial distribution of some sociological characteristic. Thus, the province of Quebec corresponds to an area in which French-speakers are a majority (despite Anglophone pockets such as West

Montreal). The Swiss cantons are divided along linguistic and religious lines. Few of the cantons practice bilingualism, and many of them are overwhelmingly either Protestant or

Catholic. Clearly, however, sociological distinctions are not necessary.

The U.S. states of Washington and Oregon provide an excellent example of the effects of federalism. On first glance, there is little to differentiate the two states. The critical historical difference, and the one that created two separate states, is that Oregon was settled earlier and thus gained statehood at a point when what became Washington was still at a embryonic stage. But in many respects, the states are quite similar.

Traditional economies in the western sections of both states focused on fishing and 57 logging, while the drier eastern sections centered on ranching, wheat and hitproduction.

Both have benefited from the explosion of high technology. Divisions within the two states, such as the diff'ering economies on the east and west sides of the Cascade

Mountains or the resentment by hinterland areas of Seattle or Portland, seem almost congruent. Yet for all the similarities, two separate political cultures have developed.

Washingtonians talking among themselves may say Oregon seems just a little different, a little "off."6 Oregon does seem to have a bit more of a Green streak to it, as well as stronger element of social activism. These may merely be differences of degree, but they are recognizable. Even more concrete are differences in tax systems. Washington is one of seven states in the U.S. not to have a state income tax, raising its finds instead through a 7.8% sales tax, higher licensing fees and other means. Oregon, on the other hand, has an income tax, but no sales tax. This is such a point of pride to some Oregonians that when traveling in Washington, they refbse to stay in hotels or motels that insist on charging them the sales tax. So, despite the similarities between the two states, federalism has permitted the creation of two distinctive political cultures and systems. It is true that rivalry and differences between Seattle and Portland are no more pronounced than those between Edmonton and Calgary, or San Francisco and Los Angeles, but by being in separate federal jurisdictions, the reflection of those differences through different laws and policies is more pronounced.

Further evidence of the emergence of differing political cultures within the U.S. is provided by Daniel Elazar. Elazar developed a taxonomy of the various political cultures 58 of the 50 states, slotting the states into three types of political cultures - individualistic,

moralistic, and traditionalistic - with some states having combinations of those types.

The three types view the democratic ideal differently. The individualistic political culture

emphasizes the marketplace, wanting a limited role for government (1972: 94). The

moralistic view strives for the "good society," seeing good government as "measured by

the degree to which it promotes the public good" and is conducted openly and honestly

(1 972: 96-97). Traditionalistic political cultures seek to maintain the position of long-

standing elites and the hierarchy on which their power is built (1972: 99). At the time he

wrote, Elazar found traditionalistic political cultures throughout the South, moralistic

cultures in New England and the Pacific states, and individualistic cultures in the

Midwest and the Plains. Considerable overlap and mixing is present, however. To return

to the example discussed above, Elazar categorizes Oregon as a moralistic political

culture. Washington is also said to have a moralistic political culture, but one with

individualistic strains.

Whether federalism is a reflection of a pre-existing societal division or a creator of one, the central point remains. Federalism provides a means for differences to be both protected and enhanced. This demonstrates Madison's point in Federalist 51 about the protection of minorities. A minority political culture which might be subsumed within a unitary state has the space in which to flourish and even grow. In fact, the presence of meaningful borders between jurisdictions can do more than merely reflect spatial differences, it can also aid in their development, as Cairns (1988) has pointed out. 59 The division into regional areas is also important to federalism's anti-majoritarian aspects. The U.S. Senate provides equal representation to states regardless of size.

Senators representing less than 25 percent of the U.S. population can vote down legislation, or conversely, pass legislation. An even smaller group can prevent legislation fiom ever coming to a vote through the filibuster device. If the issue at hand is under the jurisdiction of the sub-national units, policies which have support on a national basis may be approved by vast numbers of states, but fail to be approved in others. Near the conclusion of Federalist 10, Madison cites this as one of the great strengths of federalism

- majorities could be restrained and bad policies could be isolated in the areas of their origin.

Federalism, a federalist conception of society, envisions a multi-faceted society in which citizens have multiple linkages to government. One link is a widely shared link with the national government. A second is the link to the sub-national government - the province, state or canton. Additional links may also emerge. The European Union provides its citizens with a supra-national linkage. Local government may provide a link as well, one shared with a relatively small community. Each of these linkages is shared with different groups of differing sizes and composition. Each layer of government thus provides protection for, and even enhancement of, the identity of the community it serves.

COOPERATIVE FEDERALISM

In discussing how federalism divides tenitorial units, one major proviso must be 60 raised. As it is being practiced in a number of countries, federalism is no longer about the division of powers. At a minimum, the line between national and sub-national responsibility has become a fuzzy one. Instead, federalism has become a mechanism for service delivery, with state and local governments dispensing goods and services initially mandated or provided by the federal government.' For example, welfare benefits in the

U.S. are federal programs administered by the states. The provinces administer the

Canadian health care system. The provinces receive federal funds that supplement their own spending on health services, but in return agree to abide by the mandates of the

Canada Health Act.

The German system builds this cooperative federalism into its national institutions. The upper house of the Geman Parliament, the Bundesrat, is filled with representatives of the various Land governments. Article 50 of The Basic Law states one task of the Bundesrat is to allow the Lander to participate in the administration of federal programs.

Federal governments without this linkage provided in the German Basic Law may attempt to "buy" their way into areas of sub-national jurisdiction, even if powers are clearly delineated, by using the power of the purse to force sub-national units to carry out its guidelines. Although Canadian provinces have constitutional responsibility for health care, they cannot charge user fees, or what in the U.S. are called co-payments, lest the federal government find the province in violation of the Canada Health Act and cut off the flow of federal health fhds to the province. U.S.states are ostensibly in charge of 61 traftic and alcohol laws, but the threat of losing transportation hdskept a national speed limit in place for two decades, and led to the establishment of a national drinking age of

Such mandates challenge federalism's anti-majoritarian elements. It becomes much more difficult for a state or province to hold out against the tide of public opinion in the rest of the country. In policy areas which constitutions grant to sub-national units, national governments are still having their way. This means that a territorially constituted minority often is forced to follow the will of the national populace at the risk of losing an important stream of hding. Furthermore, the standards are not neutral. As

Frenkel(1993: 69-70) points out, national standards, particularly when based on a "right to," are, in reality, central-city standards.*

Cooperative federalism can be based on covenant. In fact, in one sense it may be seen as more covenantal, as the two levels of government truly work together, rather than operating in separate spheres. One can only take this so far, however. To the extent the sub-national units have a say in the policies, the relationship remains covenantal. When the relationship becomes one of the federal government dictating policy and forcing states or local governments to implement and, in some cases, pay for programs, the arrangement is no longer reciprocal and thus no longer covenantal. For example, in Garcia v. San

Antonio Metropolitan Transit Authority (1985) (469 U.S. 524), a 5-to-4 majority of the

U.S. Supreme Court argued, .. . we continue to recognize that the States occupy a special and specific position in our constitutional system and that the scope of Congress' authority under the Commerce Clause must reflect that position. But .the principal and basic limit on the federal commerce power is that inherent in all congressional action - the built- in restraints that our system provides through state participation in federal governmental action. The political process ensures that laws that unduly burden the States will not be promulgated. In the factual setting of these cases the internal safeguards of the political process have performed as intended.

The opinion of the Court appears to be that the covenant is limited to giving the states a

voice (through representation in the House, and in particular, the Senate) in the decisions

of Congress. Because of this representation, the Commerce Clause is essentially

unlimited and there is no sphere of state sovereignty. This can be understood as a

rejection of one of federalism's two anti-majoritarian aspects. While the majority does not have unfettered power at the center, because of the Senate, the other anti-majoritarian

aspect of federalism, taking certain issues out of the national forum, was rejected by the

Court. The Court maintained this position until 1995, when United States v. Lopez reversed Garcia in another 5-to-4 decision.

WHY FEDERALISM IS WORTH STUDYING

For all the vast range of literature which has been produced on federalism, there are voices which argue that there is little to be gained from its study. The most notable example of this comes from Rufus Davis (1978). Davis claims that the variance of federal systems, the circumstances in which they arise, the formal and informal structures which they put in place, and their practice makes comparison and generalization impossible. For example, Davis argues that Riker's necessary circumstances for the

emergence of federal systems are so broad and lacking in explanatory power as to be

irrelevant. To Davis, both a military threat and a desire for expansion are present nearly

everywhere, not just in federations. Therefore, claiming both are there

...at the birth of each federation is to state a commonplace that is hardly worth noting. It would be remarkable indeed if one could point to any of the multifoms of territorial political association throughout history and suggest that neither of these two factors were present, or indeed that the sole rationale for their origin and existence was brotherly love (Davis, 1978:137).

Riker (1970) himself later rejected the study of federalism, calling it trivial because the

circumstances of federalism and its practice were too variable and inconsistent.

One rejection of this view comes fiom Brian Galligan (1 989:3), who argues that

where federalism has been important, one cannot simply overlook it or ignore it.

.. .federalism is so pervasive in a federal country like Australia that opting o.ut is hardly feasible. Australia's political and constitutional systems, its public finance and political economy, and its political history and culture are so thoroughly federalised that it would be a superficial approach indeed that did not have a major emphasis and focus on federalism.

Galligan (1989:3-4) goes on to argue that one solution for overcoming federalism's variance is to structure research in a narrow way. Rather than seeking answers to the broad comparative questions Davis claims cannot be answered, such as "Why do federations succeed or fail?'or "Do federal systems promote or inhibit economic 64 freedom?" (Davis, 1978: 209), one can focus on areas where federalism has clearly had

an impact, such as judicial review, or the nature of federal financial arrangements. This

work does precisely that, as it attempts to identifj the characteristics of the relationship

between federalism and the use of direct democracy. That the four national jurisdictions

here all share the use of both federalism and direct democracy provides ample ground for

use of a "most-similar" systems approach (see Lijphart, 1975).

It is true, as Davis points out, that federalism has had quite a multi-faceted usage.

But the same can be said, with even more force, for populism. It is part of the scientific

endeavor to bring such concepts down to operationalizable, generalizable definitions.

When noting the impact of both federalism and direct democracy on the four countries in this study, such an endeavor cannot be seen as trivial.

' Lest this argument be seen as "too cute," the point should be raised that the New Zealand government could pass legislation for national standards on car parking, but it is more likely to delegate it to local authorities. A national government moving unilaterally on such a local issue without at least the consultation of local authorities is, at best, implausible.

Garth Stevenson (1989:7) notes that political scientists have tended toward the broader, more inclusive defdtions, whereas those with a legal background have tended to follow narrower delinitions focused on institutions.

There is a strange inconsistency here. Elazar clearly talks about covenants involving equals, but the frrst cited covenant comes fiom the Bible, a covenant between God and ancient Israel. It does not take a theologian to recognize that this is not an equal agreement.

* Jaensch (198 1:28) states this somewhat differently, saying the two spheres of power are determined and protected "by means of a judicial authority."

'To varying degrees, all four "national" states in this study also faced this task. The United States had to build a national identity that transended state identities. This process has been least notable in Switzerland, where cantonal and even communal identities remain very strong, and has achieved only minimal success in the Francophone areas of Quebec. Oregon law prohibits the pumping of one's own gas. The origins of this law are in environmental and labor concerns. Hiring people to pump gas protects jobs and also protects the environment because trained personnel, the logic goes, spill less gas than a customer would. Only one other state has such a law.

'Hence, the focus in the U.S. on "intergovernmental relations."

A Bill of Rights or Charter or Rights has a similar impact. Rights are seen as universal, thus the broader the range of rights granted within the document, the narrower the range of policy diversity sub-national units can have. CHAPTER 4 - POPULISM AND FEDERALISM

The central theme of this study is the apparent conflict between populist and federalist conceptions of the state. To be more specific, a populist conception of the state is based on rule of the general will, or at least the related, if not precisely congruent, popular will. This will is singular and indivisible; thus the powers of the state should be indivisible. A federalist conception of the state believes as a central tenet that power should be divided. Specifically, in a federal state, power is divided between the national government and sub-national units. Therefore, there appears to be an intractable conflict between populism and federalism, with the former seeking indivisible power, and the latter seeking diffuse power. However, the relationship is actually much more complex.

The next two chapters aim to explore that relationship in four ways.

First, this chapter focuses on the Federalist Papers, particularly James Madison's criticism of direct democracy and the alternatives he suggests as a remedy for the problem of "faction." Second, the chapter proposes an alternative understanding, one based on the argument that populism and federalism, at least as federalism is classically understood, can work hand-in-hand. Third, the chapter attempts to demonstrate by providing data on and discussion of direct democracy in the four countries examined here that populism and federalism have been used together in each of them.

The following chapter discusses the areas in which populism and federalism can be seen to be in conflict. The first section points out the difficulties that cooperative federalism raises for populism. Second, the chapter discusses how differing conceptions of liberty, one based on the individual, one based on the community, may come into conflict. Third, the chapter addresses the variable of size, drawing on the writings of the

Anti-Federalists as well as asking what impact size and heterogeneity of the populace have on the use of direct democracy.

MADISON'S CRITIQUE OF POPULISM

The term "populism" was not used for more than a century after James Madison contributed to the Federalist Papers. Nonetheless, Madison's scathing critique of factions and pure democracy's inability to limit them can be seen as an attack on populist conceptions of society:

As long as the reason of man continues fallible, and he is at liberty to exercise it, different opinions will be formed. .. . From the protection of different and unequal faculties of acquiring property, the possession of different degrees and kinds of property immediately results: and .. . ensues a division of the society into different interests and parties. The latent causes of faction are thus sown in the nature of man; and we see them every where brought into different degrees of activity, according to the different circumstances of civil society (Federalist 10).

Madison saw faction as an inevitable consequence of competing interests in a free society, and points to two tools for limiting the harmful effects of faction - representative democracy (a "republic") and federalism. In Federalist 10, Madison most clearly states his opposition to direct democracy.

.. . it may be concluded, that a pure Democracy, by which I mean, a Society, consisting of a small number of citizens, who assemble and administer the Government in person, can admit of no cure for the mischiefs of faction. A common passion or interest will, in almost every case, be felt by a majority of the whole; a communication and concert results fiom the form of Government itself; and there is nothing to check the inducements to sacrifice the weaker party, or an obnoxious individual. He provides amplification for this opposition in Federalist 49. In response to Jefferson, who argued disputes between branches of the Virginia govemment would best be settled by appeal to the people through the calling of a convention, Madison concedes that "a constitutional road to the decision of the people ought to be marked out and kept open, for certain great and extraordinary occasions." But in general, he finds more hazards in such a provision.

.. . as every appeal to the people would carry an implication of some defect in the government, fiequent appeals would, in great measure, deprive the government of that veneration which time bestows on everything, and without which perhaps the wisest and fieest governments possess the requisite stability.

Having established the problems of faction in Federalist 10, Madison provides a detailed explanation in Federalist 51 of how the Constitution dilutes faction's dire effects. He emphasizes two barriers to factional tyranny -the large size of the republic, and the division of powers. In Federalist 51, Madison lays out the need for a federal system in a large republic, again using as its raison d 'etre the need to overcome the problem of faction.

If a majority be united by a common interest, the rights of the minority will be insecure. There are but two methods of providing against this evil ... the second method will be exemplified in the federal republic of the United States. Whilst all authority in it will be derived from and dependent on the society, the society itself will be broken into so many parts, interests and classes of citizens, that the rights of individuals, or of the minority, will be in little danger from interested combinations of the majority.

Much of Federalist 10 and 5 1 stand in direct opposition to populist thought. Even the concept of faction runs counter to the monist basis of populism. To put it bluntly,

populist thought does not agree with Madison that faction is inevitable. The possibility

of faction is not discounted, but the good society is seen as one in which factions do not

exist, as noted by Donald MacRae (1969: 160):

Populism believes the individual should be a complete man. Complete men, living idealy in independent agrarian virtue, would agree with one another. Their insights would be sound, healthy, bound to appropriate pieties. Their judgements would be fkee but would coincide. Their society would be essentially consensual and uniform.

Although MacRae suggests this largely in an agrarian realm, the idea exists in

present-day populist movements that have moved beyond an agrarian base. Flanagan

(1995: 24-7,34-6) points out the monist thinking implicit in Preston Manning's

concept of riding a populist wave to power. Populists seek to find the popular will

and believe society can be united by a common set of interests.

As we saw in chapter two, populism places a great deal of faith in the devices of pure, or

direct, democracy. The basis of this difference between Madisonian and populist thought

can be found in two portions of the quotation fiom Federalist 10. The first is the

difference in the understanding of faction explained above. The second difference is

found in the final sentence of Madison's quotation and requires further elaboration.

Madison writes "there is nothing to check the inducements to sacrifice the weaker

party, or an obnoxious individual." Madison assumes the weaker party will be the less

numerous party, which will be outvoted at every turn. The populist critique of the

American west at the end of the 19th century found representative democracy leaving the numerically substantial farmers in a position of weakness compared to the banks and the 70 railroads. Direct democracy is seen as giving power to those who might not have the

resources to compete in a representative system. Similarly, the obnoxious individual

would likely be seen as a "self-serving elite," operating outside of, and in opposition to,

the desires of the general will. Madison assumes this person to be in a position of weakness, populism assumes this person to be in a position of strength. ' From the above discussion, it might seem as though differences between populism

and federalism are intractable. Federalism seeks to restrict majoritarianism, populism

uses it as a means of legitimation. Yet there may be room for coexistence, and for what

can be called "populist federalism." Populist federalism would accept the division of

power between the national and sub-national units, but within those units, use

majoritarian decision-making devices. To understand the underpinnings of populist

federalism, we must also recognize two different conceptions of liberty.

THE SHARED LIBERTY OF FEDERALISM AND POPULISM

Madison conceived of federalism in part as a means of protecting liberty. What

must be asked, however, is what kind of liberty? Among the modem liberal democracies,

liberty is usually conceived of with a focus on individuals. An alternative view of liberty,

and one that provides a prospective which brings populism and federalism together more

than any other theoretical perspective, is one which sees liberty as lying with

communites. As Elazar explains (1989:16), "Communal democracy begins from the theoretical premise that communties as well as individuals are of nature and the individual finds his or her rights best protected within the framework of his or her community." 71 This is best demonstrated in Switzerland. Historically, Swiss liberty is linked to

one's place in a community. Even today, Swiss citizenship is based on one having

citizenship in the commune. Liberty in traditional Switzerland was less a matter of "I

have the fieedom to," but rather, "our community has the freedom to." Benjamin Barber

(1974: 196) cites the pre-World War 11 author Edgar Bonjour:

The modem concern with the fkeedom of the individual simply would not have been understood by the traditional Swiss - so completely did they feel themselves tied to the community in every facet of their lives. They strove for independence not for themselves personally, but for the collective body. How different this is fiom the modem Enlightenment view of democratic fieedom, that disolves and atomizes the body of the state into discrete individuals.

A communal approach to liberty is consistent with populism. Rather than individual interests, there is the "common sense of the common people," which might be interpreted as "community standards." Direct democracy allows the community to state those standards with a clear voice. Federalism comes into play by protecting the diversity and autonomy of those communities. Together, populism and federalism can protect liberty, but a liberty that individuals receive through their community ties.

One must be cautious about the utility of this model. Frenkel(1989:65-6) argues that communal liberty is based in neither federalism nor democracy, but rather small- scale republicanism which emerges out of a distrust of one's neighbors. This distrust leads to avoidance of centralized power, which has, as a "happy byproduct" produced liberty. Further, one may retreat to Madison's claim that liberty is in more danger in small communities than in larger ones. The key here is to develop institutions that build participation and consensus into local governance.

Furthennore, one must question whether communal liberty can survive the 72 ascendancy of individual liberty. The social control of the community can come into

conflict with individual rights. Federalism itself is threatened by universal

understandings of rights - how can one have varying standards of services and differing laws controlling behavior if all rights are equal - especially by "positiveyyrights requiring

govemment action to provide services. A major rationale behind Trudeau's drive for a

Charter of Rights was that it would serve as a nationalizing agent that would weaken the ties of individuals to smaller loyalties and identities (Trudeau, 1993: 322-3).

Nonetheless, we can draw some conclusions here. If we accept the logic of communal liberty, and by extension, populist federalism, we can anticipate that abuses of direct democracy, where rights do get called into question, occur in less homogenous communities where any possibility of monist political culture is out of the question. It may be, then, that we will see that direct democracy becomes more divisive as the community becomes larger, meaning initiatives and referendums are more threatening in

California and Zurich than in North Dakota and Luzern.

THE CO-EXISTENCE OF POPULISM AND FEDERALISM

Populist federalism can be understood as a rejection of Madisonian democracy.

This is not surprising; any type of populism is a rejection of Madisionian democracy. But populist federalism is a also unique reversal of the thinking of Madison.

Madison seeks "bigness" to protect rights through greater diversity of interests, which prevents any interest fiom becoming the majority. Small communities are seen as dangerous for liberty because majorities are easily defined and are able to deny rights to minorities, which are equally easy to define. Populist federalism seeks to create 73 consistent majorities which can find a monist general will. Such monism is easiest to

achieve in small, homogenous communities.

Madison fbrther seeks to divide powers within levels of government, an anti-

majoritarian device that seeks to make it difficult to pass legislation. Populism rejects this. It can recognize and even support the federal division of power, because it may serve to create monist polities, but when a decision is made, it is made on a majoritarian basis.

The key to finding a linkage in the two concepts lies in one word - "people." Who constitutes the people? This is a straightforward question in a homogeneous, unitary state. But in a federal state, the question is more nuanced. Are the people the "people of the United States," or the "people of Texas"? Are they "Swiss" or Neuchatelois"? And, of course, are they "Canadian," or are they "Quebecois"?

Federal states have made this determination in either direction. In the U.S. today,

"We the people" are clearly American first, particularly since the Civil War and the ratification of the 14' Amendment in 1865. The case is equally true, but in the opposite direction, for Switzerland, where citizenship is determined at not merely the cantonal level, but at the communal level. A commune in Canton Luzern received international attention in 2000 when a number of Eastern European immigrants had citizenship applications rejected in a vote by communal citizens. However, in both countries, the less significant identity still exists and receives institutional recognition. This poses an interesting dilemna for populist thought. If populist thought is based on seeking the will of the people, or following the common sense of the common people, should a Kansas populist follow the will of Kansans, or the will of all ~mericans?~ It is here that the co-existence of populism and federalism fully emerges.

Populism understands the people to speak with one voice. Federalism permits the redefining of "people" in order to seek monist unity. Issues can be divided based on the division of federal jurisdiction. Issues of a national nature can be decided by the national

"people." Issues of a more local nature, to use the terminology of the BNA Act of 1867, are decided by sub-national "people." Who the people are, and what boundaries are placed on them, are determined by the federal division of jurisdiction within the constitution.

To what extent, then, do majoritarian direct democracy devices have a place within federal systems? Table 4.1 provides a list of the fiequency of use of referendums and initiatives within the jurisdictions in this study. As the table indicates, the most fiequent users of referendums, both of the obligatory type and those demanded by the people, are the U.S. states and the Swiss cantons, especially Zurich. While this will be discussed in more detail in chapter six, it is worth noting here that many of the obligatory referendums held in the Swiss cantons are triggered by rules requiring that loans or one- time capital expenditures over a certain cost to be referred to the voters. At the federal level in Switzerland and Australia, and at the state level in the U.S., most of the obligatory referendums are for proposed constitutional changes. Initiatives have not been present in Australia or ~anada.~

The presence of direct democracy devices indicates at least some co-existence between federalism and populism. How populist federal governance is within these systems can be determined by examining the prima facie evidence provided by the fiequency with which direct democracy is used, but also by the more subtle evidence of TABLE 4.1 FREQUENCY OF DIRECT DEMOCRACY DEVICE USAGE

JURJSDICTION REFERENDUMS FACILITATIVE INITIATIVES COUNTER- TOTAL REFERENDUMS PROPOSALS SWITZERLAND (1970-94) FEDERAL 78@ GENEVA 45 FRIBURG 47 LUZERN 27 NEUCHATEL 97 ZURICH 25 8

UNITED STATES (1 970-94) CALEORNlA 267 MONTANA 4 1 NORTH DAKOTA 78 OREGON 120$ WASHINGTON 24

AUSTRALIA (1901 -94) COMMONWEALTH 42

CANADA (1895-1 994) FEDERAL 3

@ - Eleven of the Swiss federal referendums were on non-constitutional matters. # - Includes four measures placed on the ballot by a constitutional convention. $ - Includes six advisory, non-binding referendums. * - Includes one indirect initiative approved by legislature without being sent to people. 76 the variety and ease of methods by which direct democracy is triggered, and by the issues

which ballot measures address.

Taking the least populist case first, Canada has used methods of direct democracy

relatively little. Further, all ballot measures have been referendums sent to the people by

governments. Canada has had only three national referendums, and the 10 provinces

have totaled less than 40, more than half of them on issues of either alcohol consumption

and distribution or time zones and daylight savings time (Boyer: 1992: 23 1-33). While a

handhl of referendums have been critically important in Canadian history (these will be

discussed in more detail in chapter six), the general climate toward referendums has been

hostile. In a 1916 case, a Manitoba law providing for the use of referendums was struck

down on the basis that in no regard were the people sovereign (Morton, 194: 287).

As in Canada, all Australian ballot measures have been referendums sent to the

people by government. The difference in Australia is that populism, through the

referendum device, is recognized as a legitimate force in the Australian constitution. All

proposed constitutional amendments gaining approval fiom the federal legislature must

be sent to the people for ratification. Thus, 44 constitutional referendums have been sent

to the Australian people since 1901. Referendums at the state level, however, have a

history remarkably similar to those in the Canadian provinces. State-level referendums

have been Mequent and centered largely on issues of alcohol and time zones (Hughes,

1994: 167-8).

In the United States, the constitution may begin with "We the people," but as we

have seen, Madisonian democracy as practiced in the federal govenunent rejected populist devices. At the state level, however, 49 of the 50 states send amendments of the 77 state constitution to the people for ratification, 25 have the facilitative referendum and 24 use the initiative (Book of the States, 1996-97: 5,209). The five states focused on in this study have all three elements in practice. The use of direct democracy has been extensive in four, and relatively less so in the fifth, Montana. The fiequency of ballot measures may be partially dependent on the state's political culture, but the institutional environment also plays a role. For example, one sometimes overlooked factor in the fiequency of the ballot measures, and in particular, the fiequency of referendums, is the complexity and scope of the state constitution. If the state constitution is a wide-ranging, detailed document, legislatures may be forced to resort to the referendum on a more fiequent basis to enact policies which would fall under statutory law in other jurisdictions. In California, the constitution contains some 20 articles, compared to seven in the U.S .federal document. As shown in table 4.1, California's use of referendums sent to the people by the legislature has been extensive. Of the 267 referendums, 155 were constitutional referendums, which on its own remains by far the highest number of referendums in the five states.

There exists an additional factor in the high number of obligatory referendums used in California. This is the unique entrenchment of statutory initiatives in California law. According to Art. 11, Sec. 10c of the California Constitution, the state legislature cannot overturn or amend statutory initiatives passed by the people. Any bill passed by the legislature which alters an initiative is sent to the people for final approval. This provides statutory initiatives in California with a quasi-constitutional status. Only if the new law violates some pre-existing clause of the state constitution can it be overturned, and this would occur through the courts, rather than through the legislature. This status is 78 unique in the U.S., although some less stringent protections are provided in other states.

Washington statutory initiatives have a two-year entrenched period, after which

overturning or amending the initiative requires a two-thirds majority in both chambers of the state legislature (Book of the States 1996-97: 215).

A central factor in the use of initiatives in the U.S.states is the number of signatures required for the measure to be sent to the people. The lowest threshold is in

North Dakota, where initiatives on statutes reach the ballot upon the signatures of two percent of the resident population, and constitutional initiatives reach the ballot after achieving a four-percent threshold. Those low thresholds are the most likely explanation for why North Dakota uses the initiative more than states of similar demographic characteristics or fiom the same region. The simplest comparison can be made with

South Dakota. South Dakota has thresholds of five percent for statute initiatives and 10 percent for constitutional initiatives. From introduction in 1914 through to 1994, North

Dakota had sent 160 initiatives to the ballot; South Dakota, despite a 16-year head start as the fist state to introduce the device, sent only 42 (Tolbert, Lowenstein, Donovan, 1998:

29).

While the facilitative referendum is available in all five states, its use has been more limited than that of the obligatory referendum or the initiative. In California, the facilitative initiative has not been used since 1952, except for four measures which reached the ballot in 1982, all of which failed. Lee (1978: 100) attributes this to two factors. First, the California legislature now meets on a virtually full-time basis, meaning the opportunity to amend legislation presents itself more fiequently, and second, rather than deal with the tight time limits of the facilitative referendum, groups have found they 79 can accomplish the same goals through the initiative. Washington, Oregon and Montana

have experienced similar limited use of the facilitative referendum? In North Dakota,

use has been somewhat more frequent, with 18 facilitative referendums reaching the

ballot. Still, even in North Dakota other mechanisms have been used more frequently.

In Switzerland, ballot measures are an integral part of the political system. At

both the national level and in the canton and commune (local) governments, dates for the

presentation of ballot measures to the people are scheduled every three months.

Although sometimes no measures are presented on the scheduled dates, more frequently

voters made decisions on at least one and sometimes multiple measures. For example, at

the federal level, voters went to the polls 74 times out of the 100 scheduled dates in the

25 years of 1970-94, voting on a total of 204 measures.

At the federal level, the vast majority of referendums are constitutional, save for a

handful concerning treaty ratifications, while in the cantons, more referendums deal with

expenditure limitations. Programs over a certain cost, particularly for capital spending,

must be approved by the electorate. This mechanism finds its most frequent use in

Zurich. Zurich voters voted on an average of 11 obligatory cantonal referendums a year,

both constitutional and statutory, in the period 1970-94.

One immediately evident difference between the Swiss cantons and U.S. states is

the relative preference in Switzerland for the reactive device of the facilitative referendum rather than the initiative. Only in Zurich have initiatives outnumbered facilitative referendums in the 25-year period of study, in'part because of tighter limitations on how facilitative referendums can be used, but even there, the nurnber of facilitative referendums is greater than in any of the five U.S. states. While this will be 80 dealt with more thoroughly in chapter six, it is worth noting here that Swiss legislative

sessions are very short, leaving many opportunities for the facilitative referendum.

Furthermore, the signature requirements for getting a facilitative referendum to the ballot are in some cases smaller than for initiatives. In Luzern, 4,000 signatures are required to place an initiative on the ballot, but only 3,000 are required to challenge legislation through a facilitative referendum. In Geneva, 10,000 signatures are needed for initiatives, but only 7,000 for the facilitative referendum. This is especially true at the federal level - 100,000 signatures are required for initiatives, which are always treated as constitutional amendments, and only 50,000 for the facilitative referendum. Initiatives still occurred more frequently at the federal level - 65 initiatives to 50 facilitative referendums. What is different is their success. Some 68 percent of the facilitative referendums were successfil as opposed to less than eight percent of the initiatives.

Thresholds in Switzerland are always set to a specific number, rather than the

U.S. method of setting a percentage of the eligible electorate or those actually voting in the most recent general electorate as the threshold. Table 4.2 provides those thresholds, converting the Swiss thresholds into percentages of the eligible electorate as of 1994 for easier comparison.. Overall, signature thresholds in Switzerland, at least in the five cantons examined here and at the federal level, are generally lower than in the U.S. states shown. Evexi so, Swiss levels of initiative use are substantially lower than in the U.S.

Only at the federal level and Zurich has the use of initiatives, at least in a quantitative sense, matched that of the U.S. The relative paucity of initiatives in the other four cantons is unheard of in U.S. states with low thresholds. Only in states with thresholds of

10 percent or more is the device used so infrequently. Again, it must be noted that the TABLE 4.2

BALLOT MEASURE THRESHOLDS IN THE US AND SWITZERLAND

INITIATIVE FACULITATVE STATUTORYICONSTITUTIONAL REFERENDUM

CALIFORNIA 518% MONTANA 5110% NORTH DAKOTA 215% OREGON 618% WASHINGTON 8110%

FRIBOURG 4.114.1% GENEVA 5.015.0% LUZERN 1.811.8% NEUCHATEL 5.9/5.9% ZURICH 1.3/1.3%

SWITZEFUAND ---/I. 1% 2.2%

Swiss eligible electorates: Fribourg 147,955; Geneva 201,78 1; Luzern 224,893; Neuchatel102,503 (Sept. 1993); Zurich 762,960 @ec. 1992).

* - Geneva grants only a 40-day period for gathering signatures after legislation is passed, rather than the 90 days granted by other jurisdictions.

SOURCES: Delley & Auer (1986). "Structures politiques des cantons,"; Book of the States, 1996-97: 209 U.S. states selected for this study are extensive users of ballot measures to a level

somewhat beyond that of most states. Nonetheless, it is apparent fiom the two tables that

the threshold is not the only determinant of how many initiatives reach the ballot.

In both Switzerland and the United States, a wide range of issues are presented to

the voters. Table 4.3 provides a detailed categorization of the issues presented to the

voters. Eight .categories have been created, and an "other" category has been added for

remaining issues.'

In most jurisdictions, issues of political structure are the most common items

placed before the voters. While these can become issues of importance and be hotly

contested, such as a California referendum on reapportionment that passed with 55

percent support, many more are routine measures that attract little attention, much less

opposition. A 1978 Oregon referendum providing open meeting rules for the state

legislature passed with more than 80 percent of the vote.

Previous research has shown referendums have a much higher rate of success than

initiatives (Ranney, 198 1 : 77). Magleby (1 984: 72-3) speculates possible reasons for this

difference may be because voters have greater trust in elected representatives than in

interest groups, and because a portion of the referendums are on uncontroversial matters

and sent to the voters with unanimous legislative support. The strength of support for

obligatory referendums is most evident in the Swiss cantons. In the five studied cantons,

the lowest percentage of obligatory referendums passed was 81 percent in Luzern. In

Neuchatel, the success rate was 92 percent. In the five U.S. states, rates of approval are high, although not to the Swiss extent. Oregon voters have shown the most skeptical

attitude, approving only 52 percent of obligatory referendums sent to them. TABLE 4.3

BALLOT MEASURES CATGORIZED BY TOPIC AND JURISDICTION

SWITZERLAND UNITED STATES AUSTRALIA TYPE FR GE LU NE ZH CH CA MT ND OR 'WA AU

Environmental 1 3 5 .4 27 20 44 5 4 27 16 0

Transportation 10 15 5 13 49 25 9 0 2 14 3 2

Political Structures 28 18 12 13 49 13 63 24 60 50 35 19

CulturaVMoraV 6 11 9 4 34 40 3 1 9 14 26 9 4 Rights

Education 5 5 10 22 42 8 43 3 9 22 7 0

Economic 0 8 6 5 13 17 3 0 8 17 4 5 15

Other 6 11 5 10 44 24 27 5 5 16 8 0 84 The discussion above has focused mostly on non-national level ballot measures.

From the data detailing the use of direct democracy in sub-national units of Switzerland

and the United States, it becomes apparent that some level of populist federalism is possible. Voters in the Swiss cantons and U.S. states have the ability, within their jurisdictions and the areas of competence of those jurisdictions, to use populist devices.

They have done so on a wide range of issues ranging fkom dentures to voting rights. The question remains as to whether this can also occur at the federal level. No direct democracy exists at the federal level in the U.S.,but it is critically important in

Switzerland and Australia, and has also played a role in Canada.

National referendums in Australia and Switzerland require a "double majority" for passage (Kohach, 1993: 42; Jaensch, 1981: 32). In other words, both a majority of the voters nationwide, and a majority of the voters in a majority of the states or cantons, must approvem6Thus, in Australia and Switzerland, the question of whether the sovereignty of the people lies with the federal level or the sub-federal level is answered by saying it lies with both. Galligan (1995: 114) describes this in more detail:

.. .the essential features of a federal republic are dual citizenship of the people in the two political communities, Commonwealth and State, and sovereignty of the people as the ultimate source of political authority. The Australian amending formula embodies both principles to a high degree. The federal principle is satisfied by requiring, in addition to an overall majority of the electors, majorities of electors in a majority of the States. The principle of democratic sovereignty of the people is evident in having electors themselves vote on proposed changes rather than having legislatures or ratifying conventions of elected delegates determine the outcome on the people's behalf. While the latter arrangements are perfectly compatible with representative democracy, popular referendum is more directly democratic.

In both countries, referendums or, in the Swiss case, initiatives or facilitative referendums, have been defeated despite getting a majority of the overall electorate

because they failed to receive majorities in a majority of states or cantons. In

Switzerland, eight measures have failed because they achieved one majority but not the

other (Linder, 1994: 74). The four cases since 1970 are quite similar - in each case, the

measure was a referendum put forward by the federal government attempting to broaden

federal authority in a given policy area, but all failed to receive majorities in a sufficient

number of cantons. In Australia, five constitutional referendums have received a

majority support fiom the electorate but failed to get majorities in the required four of six

states (Jaensch, 1994: 53-55). The first three were efforts to give the Commonwealth

government greater jurisdiction in a given policy area, for example, aviation in 1936.

The two most recent examples were rejections in 1977 and again in 1984 of proposals by

Labor Party governments to always hold Senate and House of Representatives elections

simultaneously.

It is not merely the double majority that appears to slow the movement of power

and resources toward the federal level. Some have argued constitutional referendums

themselves may do so. Linder claims the referendum process has raised the bar and made

constitutional change more difficult, thus protecting the autonomy of the cantons

established in Article 3 of the constitution (1994: 42-3). Crisp (1983) goes into more detail, arguing centralizing amendments are among the most likely type of proposals to be rejected, perhaps because of a vague, but thinly defined belief in "federalism," a fear of power being taken out of local hands and thus moved fbrther fiom popular control, and a general lack of knowledge about the constitution. Furthermore, referendums tend to bring out what in Switzerland is called the Neinsager vote, a segment of 10-25 percent of 86 the electorate which votes no to even the blandest proposals (Kobach, 1993: 48). Cronin

(1989:88) claims cautious voters take a status-quo, "When in doubt, vote no," approach to

ballot measures. These two factors are likely to be amplified in Australia, where voting

is compulsory, meaning those with limited understanding of the issue, or who have a generally anti-system outlook, and would be likely to stay home without compulsion, go to the polls.

Perhaps in part because of those factors, Australia is the proverbial exception to the rule about obligatory referendums tending to be approved. Only eight of the 40 referendums sent to the people have gained approval. By comparison, obligatory referendums in Switzerland in 1970-94 were approved 59 of 77 times.

In two of Canada's three national ballot measures, no double majority was required. These measures - the 1898 vote on prohibition and the 1942 vote on conscription - were not constitutional, but rather efforts by the federal government to get out of a box created by previous commitments (Boyer 1992: 16-43). In 1898, the Liberal government was carrying out a promise it had made to temperance movements while still in opposition. In 1942, the government was seeking approval to back out of a commitment not to introduce conscription for the war effort. In each case, however, a province-by-province interpretation of the measures was unavoidable, as both passed comfortably in eight provinces and were decisively rejected in only one, Quebec.

The vote on the Charlottetown Accord was a constitutional exercise. While this will be dealt with in more detail in chapter six, two points are worth noting here. First, the idea of a double majority was present, but in an even more stringent fashion, as approval required majorities in all 10 provinces because some provisions of the accord 87 involved the unanimity provision of the amending formula. Second, while the Accord

was a complex and detailed document, it is clear it was not an effort to centralize power

in the federal government. In fact, one of the criticisms marshaled against it by some

groups was that it would weaken the federal government. This means the Charlottetown

Accord was different in a crucial respect. Nonetheless, it failed to get a majority either of

the people as a whole or of the provinces.

The Canadian case may provide a partial counter-argument to what seems

apparent fiom the other three countries. It would seem those seeking to maintain the

autonomy of states, provinces and cantons would be well-advised to argue alterations in

the division of powers between national and sub-national governments should be placed

in the hands of the people, where sub-national governments have fared quite well. But

constitutional amendments are not the only method by which this matter has been

determined. The influence of the spending power on blurring lines of competency will be

discussed more filly in the next chapter. In the present context, however, the role of the

courts must also be considered. While it has not been consistent, the long-term trend line

of the impact of the courts has been one of sending more power to the center. The courts

have been central to the expansion of power of national governments in Australia and the

U.S. Switzerland, with the weakest elements of judicial review and the strongest

tradition of direct democracy, is the most decentralized of the four countries. How does

Canada, which some call the second most decentralized country in the world behind

Switzerland (Jackson, 1992), fit in here? Canada has, aside fiom 1992, avoided popular involvement in constitutional issues. Further, one can argue the courts, and prior to that the Judicial Committee of the Privy Council in London, have protected provincial 88 prerogatives. However, this may have changed with the introduction of the Charter of

Rights and Freedoms in 1982. The idea of universal rights may push Canada toward a

path more similar to the trajectory followed in the United States.

Populism, as operationalized through direct democracy, has had a presence in all

four of these federations. Ballot measures have been used in a wide range of circumstances, on a wide range of issues, with a wide range of rules governing their use, and with a wide range of frequency. Despite this diversity, some aspects appear to be consistent across the four countries. For example, the most common issue, with only some exceptions, is that of political structures and rules. Rather than any particular kind of policy issue, the voters most often face decisions about changing the process by which policies are made and the parameters within which that process occurs. This is particularly true in Australia, but is also true to a lesser degree in Switzerland and the

U.S.

Furthermore, in Australia and Switzerland, rules regarding double majorities effectively blend populist and federalist elements. With double majorities, it can be said the people have two faces - one federal and one local - and that both faces are recognized.

Chapter six will discuss in greater detail the impact of double majorities on policy.

Nonetheless, limits to this co-existence are present, and shall be examined in the next chapter.

' Of course, in referring to weaker interests, Madison did not mean financially weaker interests. Madison and other founders assumed, in the wake of Shay's Rebellion, that the weaker interests would be those who were numerically weaker but financially stronger. The populists of the late 1800s and those advocating campaign finance reform in the US. today are making the claim that representative institutions are prone to being dominated by fmancially strong interests, regardless of their numerical strength.

Furthermore, the lack of direct democracy devices at the national level in the U.S. means the people cannot speak with a direct, popular will. Everything demanded by the people is filtered, either through representation, or through the formality of the electoral college determining who will be President.

British Columbia and Saskatchewan have passed legislation providing for initiatives, but neither has yet to be used. ' How applicable Lee's reasoning is in these cases is unclear. The initiative may be used in a way similar to California, but the legislatures in Washington and Oregon qenot full-time. In Washington, for example, the legislature meets for a 120-day session in the first year after a general election and for 45 days in the second year.

It is conceded that placement within these categories is sometimes arbitrary. For example, is a vote regarding a motor fuel tax placed within the transportation category or under revenueltaxation? In such cases, taxes and other levies for specific purposes were placed within the category of the specific purpose, thus a motor fuel tax would be listed as a transportation issue. Also, the welfare aspect of the healthlwelfare category was broadly interpreted to also include housing issues such as the building of subsidized residences or the imposition or removal of rent control.

There have been a handful of exceptions, all in cases where the vote was not on a constitutional matter. Australia's 1977 vote on a national song, which had no component for separating out the states, is perhaps the best example. CHAPTER 5 - POPULISM VERSUS FEDERALISM

The previous chapter established that populism and federalism can and do co- exist. This chapter discusses the limits on that co-existence and factors which make it difficult. First, the chapter deals with the challenges raised by cooperative federalism.

Second, it raises the issue of minority rights and the problems caused by a populist conception of government. Further to this point, it considers how the internationalization of rights may preclude a diversity of policy outcomes in small jurisdictions. Finally, the chapter discusses how the size and heterogeneity of a jurisdiction play a role in how divisive the use of populist devices is. Playing a role in each of these criticisms is a rejection of the idea that federalism can effectively provide divided definitions of who the

"people" are.

COOPERATIVE FEDERALISM

If federalism is to provide multiple conceptions of the "people," it must operate in the realm of dual federalism, where decisions can be made with a large degree of independence. The problem with this understanding is that it increasingly exists only in the theoretical realm. First, the classical federalist idea of divided jurisdiction is one that is increasingly anachronistic. Lines of jurisdiction have become blurred by practices of cooperative federalism, in which two or more levels of government work together to produce and implement policy. Furthermore, federal divisions of jurisdiction are not necessarily determined by the will of the people or based on attempts to coherently sub- 9 1 divide those peoples. The discussion of Washington and Oregon in the previous chapter is a prime example. Many cantonal boundaries in Switzerland do correspond almost perfectly to religious or linguistic divisions, but others, particularly in Bern, do not. In

Canada, Quebec may be the primary home of Francophone Canadians, but significant

French populations reside just outside Quebec's borders in northeast Ontario and northwest New Brunswick, and a large Anglophone population remains in Montreal and other portions of southwest Quebec.

Populismysproblem with cooperative federalism

Cooperative federalism1 presents two problems for the coexistence of populism and federalism. The most apparent difficulty is that when the lines of jurisdiction become blurred, the lines of identity also get blurred. A less readily noticeable but equally important difficulty is that inter-governmental relations often leads to decision- making that is increasingly bureaucratic and administrative, meaning it is taken further fiom democratic accountability. These problems shall now be discussed.

In a federal system where a high level of cooperation between levels of government exists, the classical understanding of divided sovereignty is eroded. This is a modem phenomenon, one that gets its initial push &om responses to the economic downturn of the 1930s. It is present, in varying levels, in all of the states covered in this research. Wiltshire (1992: 175-6) argues that "layer-cake federalism is no longer feasible

.... Marble-cake federalism, organic federalism, or whatever term can be used to denote the breakdown of (dual) federalism, is here to stay." Cooperative federalism is in at least 92 one respect a one-way street, with virtually all the funds flowing downhill fiom the center outward. Because of its financial leverage, policy is often created and imposed by the national government with or without the consent of sub-national units. Therefore, the general trend of cooperative federalism since the Great Depression has been toward more central leadership. However, the last 10-15 years have witnessed a shift back toward decentralization in the U.S., Canada and Australia; and Switzerland still retains a greater amount of sub-national autonomy than any of those states.

Table 5.1 provides a comparison of the distribution of revenues in each country and an indication of the dependency of lower levels of government on transfers of funds fiom higher levels. Column 1 provides the total revenues received by each level of government. However, these figures are somewhat misleading, because transfer payments between levels are included here. Thus, funds transferred fiom the

Commonwealth to Tasmania would show up twice, first as money received by the

Commonwealth in taxes or similar revenue streams, then again as money received by the states, but fiom the Commoiiwealth. Therefore, Column 1, while the simplest calculation of revenue, is flawed as it overstates the actual revenue-raising ability of sub- national and local governments. Column 2 provides a more actual accurate measure of revenue-raising ability by removing transferred fhds from the revenue calculation.

Column 2 indicates the dominant position of national governments in Australia and the

United States and also of the uniquely weak position of local government in Australia.

Column 3 provides the percentage of revenue provided by transfers fiom other levels of government. In other words, it provides a quantitative measure of dependency. National TABLE 5.1

DISTRIBUTION OF TAX REVENUES AND PROGRAM SPENDING

1 2 3 4 % TOTAL % NON-TRANSFERRED % REVENUE FROM % OUTLAY REVENUE ' REVENUE TRANSFERS

AUSTRALIA 1994-95

National 57.5 72.7 Sub-national 35.0 23.6 Local 7.5 3.7

CANADA 1995-96

National 37.3 46.3 Sub-national 43.5 41.2 Local 19.2' 12.5

SWITZERLAND 199 1

National 29.8 36.1 Sub-national 40.0 37.2 Local 30.2 26.7

UNITED STATES 1995

National 48.6 61.6 Sub-national 28.0 20.9 Local 23.4 17.5

Column 1 indicates percentage of total government revenue in the country coming to each level. This includes transfer funds, which are counted multiple times. Column 2 eliminates transfers, indicating what percentage of the total public hdsraised are brought in by each level of government prior to those funds being transferred. Column 3 indicates percentage of that level's total revenue generated by transfers fiom other levels. Column 4 indicates percentage of direct outlays (i.e. hdsnot transferred to other levels).

SOURCES: Statistisches Jahrbuch der Schweiz 1994; Canada Year Book 1997; Statistical Abstract of the United States 1999; Year Book Australia 1997. governments raise virtually all of their own revenue. Other levels receive important

portions of their revenue fiom other levels. National governments provide funds to both

levels of government beneath them. Sub-national governments provide funds to local

governments. In Switzerland, revenue also moves up fkom local governments to the sub-

national cantons, a rare exception to the downhill nature of transfers. Column 4 provides

the percentage of total outlays provided by each level of government. This gives an

indication of the amount of services provided by each level, regardless of where the

initial hding came fiom.

Differences in cooperative federalism are apparent fkom this chart. Australian

federalism is the most centralized, with the state and especially the local levels highly

dependent on transfer revenue. Local government plays only a limited role in Australian

governance, providing a very small percentage of outlays and having a minimal ability to

raise its own funds. The Australian structure of fiscal federalism is perhaps most closely

matched by that of the United States. The most important difference is in the relative

autonomy of state and local governments - they are roughly half as dependent on transfer

funds as their Australian counterparts. Another difference is the relatively higher revenue

and spending power of local government in the United States. This comes mostly at the expense of the states, but should not be overstated. State governments have a great deal of influence on local governments, both as providers of revenue and through their ability to dictate local government policy, as most states do not constitutionally guarantee local government areas of jurisdiction. 95 If Australia and the United States fit together as using more centralized fiscal

federalism, Canada and Switzerland fit as a more decentralized pair. In both countries, the ability of the national and sub-national levels to raise revenue are roughly even, with the Swiss cantons actually bringing in more revenue than the federal government. Also, both the cantons and provinces are the largest provider of outlays in each country. The biggest difference is in local government. The Swiss communes raise a higher percentage of revenues, are less dependent on outside revenue, and provide more in outlays than local government in the other three countries. For more in-depth understanding, here is a brief overview of the development of cooperative federalism in each of our countries.

Canada

Canadian federalism has been characterized by an extensive amount of interaction between the federal government and the provinces. In the words of Donald Smiley

(1 987:86), "... a continuous process of federal-provincial consultation and negotiation is at the heart of the Canadian federal system." The balance of federal and provincial powers has been on a pendulum, with strength moving back and forth between the two levels.

Such a pendulum probably puts the provinces in a better position than American and

Australian states, where most of the movement has been toward increased federal power.

Canada's blend of federalism and Parliamentary institutions has led to the development of "executive federalism," a phrase Smiley coined. Cooperative federalism thus becomes a system of deals and negotiations between the executive branches of two levels of government. These deals were initially handled at the departmental level 96 through bureaucratic channels but by the 1970s had moved largely to the "first ministers"

level of the federal Prime Minister and the 10 provincial premiers.*

Since World War 11, the federal government has made its presence felt most

strongly in two areas - equalization to poorer provinces and grants for social spending.

Equalization, which in effect redistributes finds fkom richer provinces to poorer ones,

was constitutionalized in the 1982 Constitution Act. Sect. 36, Para. 2 states "Parliament

and the government of Canada are committed to the principle of making equalization

payments to ensure that provincial governments have sufficient revenues to provide

reasonably comparable levels of public services at reasonably comparable levels of

taxation." While all provinces receive finds from federal coffers, those who cannot

provide for themselves receive proportionately more. In the 1994-95 fiscal year, for

example, 45.5 percent of the revenue of the province of Newfoundland and Labrador, and

40.5 percent in Nova Scotia, came via federal goverknent transfers. At the other end, the

"have" provinces of British Columbia and Alberta received just 11.1 and 12.4 percent of their revenues, respectively, from transfers.

Nonetheless, the provinces have been able to stake out their autonomy in ways that neither U.S. nor Australian states have been able to do. Some of this impetus for a more dual federalism is provided by the province of Quebec. As the primary home of

Canada's Francophone minority, Quebec has, especially since the 1960s, sought to have control of its own social programs. For example, Quebec has a Quebec Pension Plan separate fiom the other nine provinces, which use a program run by the federal government. Having the second-most populous province making consistent demands for 97 increased autonomy has led to similar demands fiom other provinces (Gibbins, 1990:

23 1).

Compared to the U.S., "Canada has many fewer shared-cost, conditional, or grant-

in-aid programs; those it has, involve far less detailed federal control and supervision, and

a much larger proportion of federal transfers to the provinces take the form of

unconditional grants" (Simeon, 1995: 25 1). This lack of intrusion has only been

amplified in the late 1990s with the transformation of the Established Program Funding

(EPF), which provided funding for post-secondary education and health care, and the

Canada Assistance Program (CAP) into the Canada Health and Social Transfer (CHST),

which essentially provides block funding to provincial governments with few restrictive

guidelines. At this point it is unclear whether the 1999 Social Union fiarnework

agreement will accelerate or reverse this trend. On one hand, the social union gives the

provinces the ability to opt out of federal social programs and be compensated to run their

own programs. On the other, the decentralizing presence of Quebec is not involved, as the

province refused to sign the agreement, and further, the ability of the federal government

to grade the provinces on how well they are meeting national standards may provide

pressures for increased standardization.

United States

In the U.S., the federal government holds a much stronger fiscal position than that of the

central governments in Switzerland and Canada, and while the Australian government brings in a larger percentage of pre-transfer revenue, Washington also has some 98 advantages that Canberra lacks. While the U.S. federal government's share of the pie declined for much of the 1980s and early 1990s, it continued to hold leverage on the states and local governments through more than 600 grant programs.

A key difference in U.S. cooperative federalism as opposed to its practice in

Canada and Australia is the lack of formal negotiation between the federal government and the states. For two reasons, the U.S. has no equivalent to the First Ministers' or

Premiers' Conferences. First is the multitude of states. What is practical with 10 or six sub-units is not with 50. Second, the separation of powers in the U.S. system weakens the possibilities of executive federalism, and does so even on a bilateral basis between the federal government and single states. What is agreed upon by the two executives may not be acceded to by the legislatures. The relative lack of interstate channels of communication in the U.S. system means the state governments have few methods of airing their grievances.

One avenue available to the states is court action, but the U.S. Supreme Court has tended, until this decade, to be unsympathetic. As mentioned earlier, from 1937 to 1995, the court interpreted the Interstate Commerce Clause of the Constitution in a broad manner, asserting in the 1985 case Garcia v. San Antonio Metropolitan Transit Authority that there were no areas of state jurisdiction the federal government could not regulate, a position which saw its first weakening in United States v. Lopez (1995). Furthermore, until 1992, the court gave wide-ranging power to the federal government to use "coercive standards," which forced the states to adopt federal guidelines or face serious funding cutbacks (Tolley & Wallin, 1995: 76-77). Australia

Australian federalism, like federalism in Canada, is characterized by a great deal

of executive federalism, although in Australia executive federalism focuses far less on mega-constitutional issues. Saunders (1990: 40) claims "Parliaments and the public probably know less about intergovernmental activity than any other category of govenunental endeavor." Inter-governmental relations in Australia are the most institutionalized of the four countries, with the Commonwealth taking an increasingly important role. Almost fiom 1901, the autonomy of the states has been threatened and has, over time, eroded (Else-Mitchell, 1982: 102). Today, the federal govemment has many fiscal levers at its disposal.

Some of these levers are constitutionally granted. Sect. 96 of the Australian constitution gives the Commonwealth government the power to "grant financial assistance to any state on such terms and conditions as Parliament sees fit" (Lucy, 1992:

296). This enables the federal govemment to step into areas that otherwise would be granted to the states.' Furthennore, since the Engineers' Case in 1920, and with only occasional exceptions, the Australian High Court has given steadily more power to the federal govemment in a wide range of areas. Essentially, the High Court ruled the

Commonwealth has sweeping power to carry out the enumerated powers granted in Sect.

5 1, declaring "federal powers would be given their full interpretation without regard to what might be left within the exclusive area of power belonging to the states" (Zines,

1990: 22). 100

One critical component of the federal power lies in the income tax. During World

War II, the federal govkmment raised income taxes substantially and offered to reimburse

the states with grants if they did away with their income taxes. This action was upheld by

the High Court in the Uniform Tax Case (1942). During the Menzies govemment of the

early 1950s and the Fraser govemment of the early 1990s, discussions were held

regarding returning some -income taxing powers to the states, but in both instances the

states, fearing a reduction of revenue, declined the offer (Lucy, 1992: 297-9).

Furthermore, the High Court has put severe limits on the ability of the states to impose consumption or sales taxes (Galligan & Walsh, 1990:7).

The most unique element of the Australian system of cooperative federalism is the

Loan Council, which institutionalizes executive federalism for the purposes of coordinating bonrowing. Agreed to in late 1927 and approved in a constitutional referendum in 1928, the Loan Council, which granted two votes to the Commonwealth and one each to the states, with ties going to the Commonwealth side, would determine the total amount of loans required by the seven governments and have that total borrowed, then distributed, by the Commonwealth. In practice, the body has been a highly malleable and adaptive organization, where agreed-upon practices have subsumed formal rules. First, the voting procedure was overwhelmed by the power and influence of the Commonwealth govemment. Meetings had become so dominated by the

Commonwealth that the states often did not receive proposals until hours before the conferences and then on what was, for all intents and purposes, a "take-it-or-leave-it" basis (Painter, 1996: 104). More recently, and counteracting the fbtchange, the states I 101 have found an increasing variety of methods to work their way around the constrictures of the Loan Council and get their own loans. This has occurred to such a degree that by the late 1980s, the Loan Council was not actually allocating any loans (Saunders, 1990).

Finally, Australia, like Canada, has used the fiscal power of the federal govenunent to provide "horizontal equalization." Through the Commonwealth Grants

Commission, grants are distributed "which (enable) each state, using comparable revenue effort, to deliver services to a standard not appreciably different fiom other states"

(Galligan & Walsh, 1990: 12-13). This occurs through two types of programs - specific purpose grants and general purpose grants. The former allow the Commonwealth to move into areas not actually enumerated in Sect. 5 1 and can turn the accepting state into

"little more than an administrative agency of the Commonwealth as far as that project were concerned," although in practice the requirements have not been quite that stringent

(Lucy, 1992: 298-9). There is, however, a real incentive to opting out of such programs, as states typically receive additional general purpose grants if they have rejected specific purpose projects, an arrangement which gives the states even more autonomy than do the

"opt out" procedures used in Canada.

Switzerland

The Swiss provide an interesting counterexample to the other countries. As with the U.S., but unlike Canada or Australia, there is little executive federalism. But unlike the U.S., Switzerland has remained highly decentralized, at a level equal to or surpassing

Canada. The revenue raising capabilities of the sub-national levels of government, not only the cantons but also the more than 3,000 local communes, far exceed those of any of

the other three countries. As indicated in Table 5.1, the communes spend a far larger proportion of public revenues than their counterparts in the other three countries, and raise a far larger proportion of those revenues on their own. This will be discussed more fully in the next chapter.

The nature of the transfer system in Switzerland differs in two significant respects fiom the other three countries. First, the federal government is not the primary sender of transferred funds, the cantons are. Second, transfers between the cantons and the communes are a two-way street (Linder, 1994). Cantons send twice as much down as the communes send back, but nonetheless, the communes are unique among the eight sub- national levels in the four states in the amount of funds they send up the system.

The federal government does have some avenues for sending money to the cantons and communes. A 1977 referendum placed the principle of cantonal equalization in the Swiss constitution, although with the rather soft language of "the Confederation shall encourage." Each canton receives a general-purpose grant fiom the Bund which operates in a manner similar to those in Australia and Canada. Further federal monies to the cantons are for specific programs. For example, the seven cantons which have universities receive additional funding.

To summarize, cooperative federalism in various forms has become increasingly relevant in all four countries. One should not see this as the post-mortem of some earlier golden era of dual federalism, however. In each country, dual federalism never existed in a tmly pure form, the form sometimes described in Canadian literature as "watertight 103 compartments." Nonetheless, relationships which have emphasized intergovernmental dealings, whether those be cooperative or competitive, have become the norm in the present time. As the lines of jurisdiction become blurred, so too does the ability to achieve the populist call of the will of the people. To the extent that money flowing fiom the center is used solely for the purpose of horizontal fiscal equalization, to use the

Australian term, and does not impose policies and standards on lower levels, jurisdiction can remain clear. But with federal fiscal dominance, national governments have sought to impose standards on their sub-units. The sub-national government is forced to cany out the imperatives of the national government, whether or not those are consistent with the imperatives of the local populace. When they do so in areas which de jure are of sub- national responsibility, national governments may be seen as an outside force impeding the "will of the people." The response may be that the federal level is also acting in the

"will of the people." However, if, as argued previously, federal divisions of power can work hand-in-hand with populism by dividing the "people" into manageable, more coherent groups for some matters, the presence of cooperative federalism, of shared-cost programs or specific-purpose grants in areas of sub-national jurisdiction violates the initial federal covenant .'

This also calls into question the democratic accountability of these dealings. One need not delve into the extensive literature on bureaucratic behavior to recognize the difficulties of unmasking and bringing into the open department-to-department level negotiations. As Ed Black (1989: 350) described the situation in Canada, even after intergovernmental power had shifted fiom bureaucrats to first ministers, "...both the mass 104 media and their audiences have been left unaware of most of the horse-trading going on between and among governments in this country." In each of these systems, negotiations between levels of government over the sharing of resources and the blending of power occur outside of public view, either between bureaucrats or in closed-door meetings between premiers and prime ministers. Even negotiations between elected politicians at the executive level tend to be characterized by a lack of legislative consultation. To the populist, such negotiation fairly reeks of elite-driven, self-interested behavior. As

Shaman (1999) points out "shifting revenue between governments can generate rent seeking and the distortion of priorities so that a state government may be forced into following a policy that contradicts the clearly expressed wishes of its state community."

Only in Switzerland, where the possibility of legislation being subject to a popular referendum is present, is there any real popular input into the negotiations, and then only as a post-hoc, negative check. Thus, both in substance and in process, cooperative federalism is inconsistent with populist principles,

In light of this, and in light of the expansion and dominance of cooperative federalism over dual federalism, one must raise the question of how plausible populist federalism is. Cooperative federalism may act in the interests of "the people" as defined nationally, but this type of understanding leaves federalism as an empty vessel, one that provides little more than improved efficiency in service delivery. MINORITY RIGHTS

As discussed in chapter two, a major objection to the use of populist devices has

been the perception of permitting a "tyranny of the majority" to run roughshod over the

rights of minorities. Data and research on this point are mixed. Some point to high-

profile initiatives targeting a given group as evidence that direct democracy is a negative,

divisive force. Gamble (1 998), for example, indicates initiatives that attack minorities

have much more success than initiatives as a whole. Implicit in such a view is the belief

that legislatures, with their greater deliberative ability, do a better job of safeguarding

minorities. Others say the record is more muddled. Donovan and Bowler (1998: 17-18)

provide the most nuanced view, claiming that the record of initiatives in the U.S. is not

actually any more hostile to minority interests than that of legislatures, and anti-minority

initiatives that have passed have often been struck down in the courts. They add,

however, that there is more hostility to minority interests in jurisdictions where those

issues have been brought to the voters.

~ederaliim,which according to Madison served to protect minorities, has also

been criticized for being a device to harm minorities. Riker (1964: 155) argued those in

the U.S.who supported segregation and racism would also approve of federalism,

because the diversity of law permitted in federalism had allowed state and local

governments to maintain racial segregation. To Riker (1964: 143) "federalism that grants

more local autonomy than is necessary for fieedom and civil liberty encourages local

tyranny, even when fieedom is narrowly interpreted as the grant of the right to minorities to have a chance to become majorities." 106 This power is not unlimited. In the U.S., the eventual willingness of the courts to

nationalize the amendments in the Bill of Rights through the 14~amendment, and of the

federal executive branch to enforce that nationalization, led to the end of the Jim Crow

laws. In Canada, the Charter of Rights has had a similar nationalizing impact. One can

recognize this was anticipated by provincial premiers through their demand for the notwithstanding clause, and in Quebec's rejection of the 1982 Constitution Act and its subsequent demands for constitutional status that would soften that nationalizing effect.

The movement toward a rights-based political discourse is not contained within national borders. Some national states have found their latitude to act being limited by outside forces. British Ministry of Defense regulations banning homosexuals fiom serving in the military were found to be violations of the right to privacy by the European

Court of Human Rights (Evans, 1999). Even national soccer leagues in Europe have been impacted by the European Court of Justice, which in the Bosman case of 1995 threw out the limitation of three foreign players on the field for all clubs. The limit was ruled a restriction on the fiee movement of labor guaranteed by the Treaty of Rome (Hughes,

1999). In these cases, the rules were entered into voluntarily by national states. The

British government was not compelled to act on the ruling of the Court of Human Rights, although it did take the immediate step of halting removals of homosexuals fiom the anned services. European Court of Justice decisions are binding on EU members, but membership in the EU itself, at least initially, was also a voluntary choice. In the spring of 1999, however, NATO bombed Yugoslavia for Serbian actions in Kosovo, indicating certain standards may be imposed forcefilly !?om the outside. 107 All this points to an internationalization of rights. One can trace this trend to the

late 1940s and the introduction of the United Nations Declaration of Universal Human

Rights. Cairns (1992: 27-30) argues this was an often overlooked factor in the push for a

Charter of Rights in Canada. Putting this into a more general context, Cairns (1992:29)

writes,

.. .the direct and indirect proselytizing on behalf of rights by the United Nations challenged regimes practicing federalism and employing parliamentary supremacy to modjfjr their constitutional mangements, as a Bill of Rights became an almost essential attribute of contemporary statehood. Accordingly, it is not surprising that a Bill of Rights has become virtually an automatic component of new constitutions, or that Bills of Rights have become increasingly comprehensive, or that an established state such as Canada, that had long existed without an entrenched Charter, has recently introduced one, or that New Zealand is seriously consid&ng doing so.

This internationalization of rights, and the concurrent expansion of the breadth of rights

to be guaranteed, narrows the range of policy discretion which governments possess.

Further, if as some have argued above, direct democracy tends to threaten minority rights,

then it becomes increasingly problematic and more difficult to justify. One can see

evidence of this in the expansion of the fkanchise to women in Switzerland, which did not

occur until 1971 at the federal level and as late as 1992 in some small Alpine cantons. In

the latter, the matter was essentially imposed upon the cantons by federal courts claiming

to be overturning an egregious violation of standard understandings of fundamental rights.

What, then, does the record show about direct democracy and the rights of minorities in the jurisdictions examined in this study? Has there been an effort to use 108 ballot measures to restrict rights or, in a more defensive action, an attempt to limit their

expansion? For this discussion, the focus shall be on the United States and Switzerland,

where the initiative process is in place. The limited number of ballot measures in

Canada and Australia require that they be left aside.6 It is conceded that the methodology

here is of limited utility, as it does not examine measures which were put forward through

regular legislative channels. Nonetheless, by examining ballot measures involving both

expansion and narrowing of rights, some sense of the general direction of the impact of

direct democracy can be established.

In Table 5.2, "threatening" measures are defined as those targeting an identifiable

group and seeking to limit it in some way. Such measures could seek to impose

discrimination on a given group, such as the 1978 California proposal which sought to

ban homosexuals fiom teaching in public school, or could seek to ban so-called positive

discrimination by removing a group's protected status. "Expansion" measures are defined

as those seeking to expand rights, such as the franchise, or to provide protected legal

status.

The table indicates voters have not uniformly rejected minority rights.

Furthennore, they have been willing on many occasions to expand those rights.

Threatening measures directed toward minority groups simply have not been a part of the experience in the Swiss cantons, at least not in the 25-year span studied. Expansion measures are as likely to be approved as rejected. In North Dakota, no issues involving minorities reached the ballot, and no threatening measures reached the ballot in Montana.

Overall, the most striking finding here is the small number of total issues put before the TABLE 5.2

RESULTS OF BALLOT MEASURES INVOLVING MINORITY RIGHTS (1 970-1994)

'THREATENING" "EXPANSION TOTAL

UNIT ACCEPT REJECT ACCEPT REJECT PRO ANTI

SWITZERLAND

FEDERAL 1 7 7 4 14 5 GENEVA 0 0 2 2 2 2 FRIBOURG 0 0 3 0 3 0 LUZERN 0 0 3 2 3 2 NEUCHATEL 0 0 1 1 1 1 ZURICH 0 0 2 3 2 3

UNITED STATES

CALIFORNIA MONTANA NORTH DAKOTA OREGON WASHINGTON voters. Only in California and in Switzerland (at the federal level) have minority issues

reached the ballot as often as an average of once every three years.

THE ISSUE OF SIZE

Most analysts argue that the rights of minorities are likely to be more secure in

larger jurisdictions. As Donovan and Bowler (1998: 1023) point out, both theory and

evidence point to minority rights being in more danger in smaller, more homogeneous

jurisdictions and ballot measures having a greater likelihood of passage. As mentioned

previously, Madison argued in both Federalist 10 and 5 1 that a larger society would offer

more protection for minorities because of the multiplicity of interests within larger

boundaries.

However, the opponents of Madison in the constitutional debates, the Anti-

Federalists, argued there was virtue in smaller entities. Storing (198 1: 15) provides three

arguments Anti-Federalists made in favor of "small republics:"

a Only a small republic can enjoy a voluntary attachment of the people to the

government and a voluntary obedience to the laws.

Only a small government can secure a genuine responsibility of the government to the

people.

0 Only a small republic can form the kind of citizens who will maintain republican

government. 111 Central to these arguments was the belief that representation in the House of

Representatives would be too limited to be effective, as the members would be

representing districts that were too large and too diverse. Brutus provides an excellent

example.

In a fkee republic, although all laws are derived fkom the consent of the people, yet the people do not declare their consent in person, but by representatives, chosen by them, who are supposed to know the minds of their constituents, and to be possessed of the integrity to declare this mind. ...Now, in a large extended country, it is impossible to have representation, possessing the sentiments, and of integrity, to declare the minds of the people, without having it so numerous and unwieldly, as to be subject in great measure to the inconveniency of a democratic government (Brutus I).

A broader discussion of the relationship of size and democracy is provided by Dahl and

Tufte (1973). They conclude that there is no optimal size for democratic governance, finding that both large and small units provide a series of tradeoffs. To summarize, small units provide effectiveness of action for individual citizens, at least where they are part of the majority, but less capacity for the system to cany out demands, whereas large units provide greater system capacity but less citizen effectiveness @ah1 & Tufte, 1973: 138).

Nonetheless, some of what Dahl and Tufte say about the strengths of small units is consistent with the sort of arguments made by populists.

Citizen effectiveness evidently depends on different techniques in the politics of homogeneity of the small system and in the politics of diversity of the large system. In the small democratic system, the chances for citizens to be effective are enhanced by the lower costs of direct communication with representatives and other officials, and by greater homogeneity, which means that even without communication representatives are more likely to hold views like those of their constituents. The large democratic system loses these advantages, as we have seen; it depends more heavily on indirect chains of communication and on overt 112 competition among organized political forces, particularly among parties @ah1 and Tufte: 1973: 109).

The value of the small jurisdiction to the populist should be quite apparent. Not only

does homogeneity provide greater opportunity for a "monist" general will, the smaller size also reduces the need for intermediary institutions such as parties, allowing citizens to have a more direct say.

The theory of populist federalism, then, calls for smaller, more homogeneous jurisdictions as being better able to establish a monist general will. To some degree, communal liberty makes similar claims. It can be argued that smaller, more homogenous jurisdictions, while perhaps being more intolerant, are also less likely to use the initiative to restrict minority rights. This is based on two factors. First, if the jurisdiction is truly intolerant in a homogenous way, it is more likely legislatures would carry out the restrictions, as representatives would be more likely to hold similar views and carry them out in office. As pointed out in chapter two, the demand for the use of direct ballot measures increases when results from legislatures seem unsatisfactory. But second, the desire for such restrictions is not high because the perceived threat from minorities when they make up a very small portion of the population is not as great. Only when the jurisdiction becomes more cosmopolitan and power becomes more contested do groups turn to ballot measures to restrict rights. If this holds to be correct, "threatening" ballot measures will be brought to the people less often in smaller, more homogeneous jurisdictions. An attempt was made to prove this statistically. The 11 jurisdictions were considered in four categories - population, percentage of the population belonging to the dominant group (whites in the U.S. states, the dominant linguistic group in Switzerland), the overall number of ballot measures 1970-1994, and the number of minority rights ballot measures in that period. The relationships did move in the direction anticipated, with larger, more diverse jurisdictions having a greater fkequency of minority rights ballot measures. Because California is much larger and is also the most diverse of the 1 1 jurisdictions, the regressions were rerun without it. Again, strong relationships in the anticpated direction were present.'

In the smallest, most homogeneous jurisdictions, threatening initiatives have been rare. It is noteworthy that in the smallest of the 1 1 jurisdictions in terns of population,

Neuchatel, the extension of political rights has been impressive. Since 1850, resident aliens have had the fianchise for cantonal and communal elections, including votes on ballot measures (Linder, 1994: 93). On the matter of granting the fianchise to women,

Neuchatel did so in 1959, becoming only the second canton to do so, and it was the first to have a woman elected to the cantonal legislature (Codding, 1961: 59). Furthermore,

Neuchatel has been one of the least supportive cantons in regards to federal ballot measures seeking restrictions on foreigners and resident aliens.' As mentioned above, the

U.S. jurisdiction with the smallest population, North Dakota, did not place a measure on the ballot involving minority rights through either initiative or referendum.

It should be noted that the relationship of size to number of ballot measures might be an artifact of the research design. It is possible that still smaller U.S. states, and in particular, smaller Swiss cantons, may have different results. Nonetheless, the sample

here does provide some indication that size is not a critical factor in the frequency of

either threatening measures or measures regarding minority rights generally. Further, the

total lack of threatening measures in seven of the 11 jurisdictions makes true statistical

analysis difficult.

CONCLUSION

This chapter has considered two problems with the co-existence of populism and

federalism. In the area of fiscal federalism, it is clear the federal spending power has

eroded the dual federalism needed for true co-existence. This has occurred to varying

degrees in the four states, and can be seen both in the percentage of revenue which flows

fiom one level of government to another and in the mandates placed on sub-national

levels of government. The expansion of a rights discourse and the internationalization of

rights have also cut into the latitude of smaller communities to act autonomously. It is

less clear, however, that the use of direct democracy, particularly in small communities,

has actually produced hostile outcomes for minority groups. Research is mixed on the

issue. The data presented here indicate such issues come to the ballot infrequently, and

do so most often in larger, more diverse communities. Chapter four demonstrates that other issues, such as taxation and structural change, are much more likely to be put to the people.

Further, while no conclusions can be definitively made here about the impact of size on minority rights, directional indicators seem to indicate smallerjurisdictions do not use ballot measures to threaten minority rights more often than larger, more diverse

jurisdictions. It is conceded that the small number of cases makes it difficult to stand on

that statement with total confidence. Additionally, this study does not delve into such

issues at the local level. Donovan and Bowler (1998: 266) argue the logic of Madison's

argument indicates "local direct democracy can be far more injurious to minorities than

direct democracy at the state level, since states are typically larger and more diverse than

lo~alities."~If local governments have some elements which change the nature of

governance and minority rights, those elements are not accessible in this study.

There is, however, a broader point of contention between populism and federalism

that emerges out of this chapter, one that focuses on the first two sections of this chapter.

It has been stated that populism is suspicious of elite-driven, closed-door processes of

decision-making. But these are the lifeblood of federalism. Federalism's lines of

jurisdiction are created and maintained primarily through two processes - negotiation

through executive federalism, and adjudication through the courts. Both are designed to

be largely outside the direct line of popular control. Except in Switzerland, the only

popular involvement occurs after executives have determined what the public will vote

on, and then typically only in constitutional matters. The presence of the initiative at the

federal level does provide the Swiss with a slightly higher degree of involvement.

Nonetheless, federalism's ability to define the various meanings of "people" is usually based on processes populists would view with suspicion. Perhaps in cases where jurisdictional lines are well settled, this is less of a problem, but these lines have rarely been static in the four countries presented in this study. "Cooperative federalism" is used here in the broadest sense, as being congruent with and including "fiscal federalism." It can be defined as any activity which uses a process of intergovernmental relations to develop or implement policies which cross lines of jurisdiction.

It should be noted that the most contentious of all matters for First Ministers' Conferences have been constitutional issues. While some issues of fiscal federalism have been constitutional, it can be argued that these conferences have had more success generally with non-constitutional issues of spending, contentious though they might be, than with constitutional issues, where the record has been one of near-total failure, with even the 1982 Constitution Act leaving out Quebec.

Sect. 51 of the constitution grants the states all "residual" powers. All powers not expressly given to the federal government are given to the states.

'For fiuther details, see Garth Stevenson (1989)' "The Origins of Cooperative Federalism," in David P. Shugarman & Reg Whitaker (eds.), Federalism and Political Community: Essays in Honour of Donald Smiley (Peterbourough: Broadview); Daniel J. Elazar (1962) The American Partnership (Chicago: University of Chicago).; and George A. Codding (1961) The Federal Government of Switzerland (Cambridge, Mass.: Riverside).

Of course, a state or provincial govenunent receiving large amounts of money from a national government may wish to maintain the status quo rather than fight for greater autonomy of action. Sharman (1999) makes this point hi a slightly different context in discussing the difficulties of meaningful tax reform in Australia.

One might argue the 1942 Conscription plebiscite was an issue of minority rights, but this is only partially coxrect. It is more accurately portrayed as an issue over which Canada's two linguistic groups disagreed. One Australian referendum which dealt directly with minority issues was the 1967 referendum on granting the Commonwealth power to pass legislation regarding Aboriginal people, a power previously held to the states. This passed with more than 90 percent support. A 1946 referendum which included similar powers, along with 13 others, to be given to the Commonwealth for a five-year "post-war reconstruction" period, failed (Jaensch, 1994: 54-5).

' Population-to-frequency of minority rights ballot measures - r=.619; percentage of majority population- to-frequency of minority rights measures - ~~562.Without California, r=.697 and -.486.

In such federal elections, resident aliens would not be permitted to vote.

It could be argued the smaller Swiss cantons are much like U.S. counties, or in the case of Geneva, a mid- sized city. In population, Geneva is roughly the same size as the metro areas of Spokane, Wash., or Madison, Wisc. CHAPTER 6 - NATIONAL DISTINCTIVENESS AND CASE STUDIES

The discussion of the last two chapters has been broad and generally comparative in scope. The aim of this chapter is to provide greater depth for the jurisdictions involved in this study, describing distinctive elements of the political systems and cultures that may impact upon the relationship between populism and federalism. The most extensive analysis will involve Switzerland, both because the nuances of the Swiss polity are not widely discussed in North America and because the Swiss have the most distinctive system of the four. Australia will also be considered in some detail. The focus on

Canada and the United States will be somewhat narrower, touching on a few key cases.

SWITZERLAND

Any discussion of direct democracy and federalism must include Switzerland as a central case. The Swiss are the most frequent users of referendums and initiatives, both at the national and sub-national level, and the possessors of, by most measures, the most decentralized federal system in the world. Some clauses of the Swiss constitution even link the two concepts. Sect. 6 states that all cantons must have constitutions in which amendments are approved by a direct vote of the people, whether that be through referendum, initiative, or through the annual landsgeminde, an open-air assembly of all citizens still held in five of the smaller cantons. The Swiss system of democracy is unique in many respects, both in institutions and practices. Therefore, a more extensive discussion is in order. Switzerland is comprised of 26 cantons, six of which are designated as half-

cantons. The half-cantons have full powers in every way, with two representational

exceptions noted below. The Federal Assembly has two houses - a lower house, the

Nationalrat (National Council); and an upper house, the Standerat (Council of States).

The Nationalrat has 200 members with cantons having representation proportional to

their population. Zurich, the largest canton with more than one million residents, has 35

representatives, while the five smallest cantons, each with populations of less than

50,000, have only one each. Members are elected through a modified dYHondtsystem of

proportional representation (Steinberg, 1996:75-76), but rather than a national list, each

canton selects its members separately.' The Standerat has 46 members, two from each of

the "full" cantons, one each fiom the six half-cantons. As in the United States, election is

for a fixed term. Each chamber is elected simultaneously for four years. The two

chambers are equal in all respects. Unlike the upper chambers in the other three

countries, the Standerat has the ability to introduce money bills. Initial consideration of

proposals is negotiated by the presidents of the two chambers (Codding, 1961: 82-3).

The executive itself is, along with the use of direct democracy, the most unique

feature of the Swiss system. The head of the Swiss executive branch is the seven- member Federal Council. Its members are elected through a joint sitting of the Federal

Assembly, with the elections the first order of business following general elections. The executive is collegial, with each member having jurisdiction over a set of departments, but none having a greater role than the others. For ceremonial duties, a President and

Vice-President serve one-year terms on a rotating basis. 119 Membership in the federal council reflects Swiss diversity in a variety of ways.

Since 1959, the four largest parties have divided the seats among them. The Free

Democratic Party, the Christian Democratic Party and the Social Democratic Party,

which each received roughly 20 percent of the vote in elections fkorn 1959 to 1995, have

two seats on the Federal Council; the Swiss People's Party, which received 10-15 percent

of the vote and is located predominantly in German-speaking, Protestant areas, has one

seat.2 This 2-2-2-1 arrangement has been referred to as the "magic formula." Further

representational guidelines are also followed (Jahrbuch, 1994:367). First, convention

holds that at least two members of the Federal Council be non-German speakers.

Convention also holds that the three largest cantons - Bern, Vaud and Zurich - are usually

represented. There is also some balance between major religious groups, as the Christian

Democrats are primarily Catholic, and the Swiss People's Party and to a lesser extent, the

Free Democrats, are primarily Protestant (Kerr, 1974: 12).

The delicate balance of the magic formula was put under considerable pressure following the elections of October 1999. After nearly a half-century of essentially stable election results, the Swiss People's Party surged to second place in seats, surpassing the

Christian Democrats and Free democrat^.^ Elements of the Swiss People's Party, easily the most conservative of the four parties, immediately demanded a second seat on the

Federal Council, although to this point, no change has been made.

The executive is elected by the Federal Assembly, but is not responsible to it.

Federal Councillors cannot be removed or compelled to resign during their term. A vote of "no confidence" does not exist. In fact, Swiss practice avoids removing executives fiom office even when their terms expire. As Steinberg (1996:137) writes, "It is

considered a great insult in most communes, cahtons and federal authorities to fail to re- elect a member of the executive. ...The Swiss voters have to have unusual provocation to let a sitting member of the executive fall fiom grace." Those whose re-election is in question, particularly at the federal level, are often ''persuaded" not to run (Hughes, 1962:

80).

The collegial-style executive is the standard throughout the Swiss political system. Executives at the cantonal and communal level have multiple members, and most cantons allocate these positions on a "grand coalition" basis, although the parties involved vary. The major exception to this would be in some highly homogeneous (and lightly populated) cantons and communes, where there is total one-party dominance with

90 percent or more backing the same party.4

The nature of the relationship between the executive and the legislature means that party discipline is weaker in Switzerland than in a Parliamentary system. Party unity scores in Switzerland are much lower than in most European countries, although somewhat higher than the U.S. (Steiner, 1974: 69-70;Kobach, 1994: 168). The lack of discipline is enhanced by two further factors - first, that the base of support is cantonal and there is not a national party or party leader dictating policy, and second, that the people hold the ability to overturn legislation, as will be further explained below. Article

3 of the constitution, which guarantees the sovereignty of the cantons, is taken seriously in Switzerland. Federalism in Switzerland

As demonstrated in the discussion of fiscal federalism in chapter four, Switzerland is the world's most decentralized federation. The Swiss system is not merely a two-level federal system, but truly a three-level system. As shown in the previous chapter, the nearly 3,000 communes have a spending power far exceeding that of local units of government in the other three countries. Linder (1 994:40) claims communes "are considered to be the foundation stone of political life and culture."

Two points demonstrate the power and significance of the communes. First,

Swiss citizenship is determined through the commune. Through having communal citizenship, one also receives Swiss citizenship. Further, this citizenship emerges not fiom the commune in which one resides, but fiom one's "commune of originy',or buergerdort, which is passed down through one's family even if they have been away for generations. Only a third of Swiss citizens live in their commune of origin, yet its importance is such that the Swiss statistical yearbook provides a map which shows the percentage of residents in each commune who hold buergerdort status in it (Steinberg,

1995: 80; Statistisches Jahrbuch, Table 19: 1).

Second, the communes have the ability to tax income (Linder, 1994: 50).' Thus, the nearly 3,000 communes, with a mean population of about 2,200, have a power that few jurisdictions that size have, and that even Australian states have been without for more than 50 years. This is key to communal autonomy - while cooperative federalism exists in Switzerland, it is not nearly the coercive force it is in the other three countries.

Size may prevent a lone commune iiom being able to provide certain services in a cost- 122 efficent way, but the resources of a group of communes are sufEcient to do so in a way that a group of local governments, even with a massive population base, would not have the revenue to attempt. The federal government does step in, providing the rough equivalent of Canadian equalization payments to the cantons, with at least some intent of taking services to people rather than forcing people to chase services and employment and thus be forced into the larger urban centers.

The decentralized nature of Swiss politics is an outgrowth of the development of the Swiss state. While all four states in this study were formed &om the joining together of smaller, pre-existing jurisdictions, Switzerland differs in two important respects.

First, the other three federations emerged fiom the linkage of colonies, whereas

Switzerland developed as a defensive mechanism to protect small jurisdictions already independent and seeking to maintain that independence. Therefore, preexisting autonomy was greater and the desire to maintain autonomy was also greater. Adding to this are the sociological differences cantonal boundaries often define. Maps providing a spatial description of religion in Switzerland (Statistisches Jahrbuch, Map 1623) show how stark the divide can be, with every commune on one side of the line being strongly Catholic and every commune on the other strongly Protestant. A similar, although somewhat less stark, situation is true linguistically (Statistisches Jahrbuch, Map 16.2).

Second, the physical distance between cantonal centers of power is much smaller in Switzerland. Only in New England and the Maritimes are state or provincial capitals in anythng near the close proximity of cantonal capitals. A 45-minute drive or train ride can easily take one through the heart of three or four cantons. Such proximity may 123 develop close relations, but it can also develop suspicions, which in turn can lead to a jealous protection of autonomy. As Frenke (1993: 65) points out, the Swiss often have a

greater dislike of neighbors in the next commune or canton than they do of those who are

Weraway.

It may be possible that the small geographic size of Switzerland has an impact on

federal legislative representatives. Members of the Nationalrat and the Standerat have a close connection to their cantons. One does not find in Swiss literature a discussion of

"inside the Beltway" culture or of being changed by living in Bern. Three intertwined factors contribute to this. One is that "home" is so close. A Member of Parliament fiom

British Columbia or Alberta goes to Ottawa and must make a fairly extensive journey to retum home. Similarly for Congressional members fiom the Pacific coast; Washington,

D.C., is far, far way. Those representing Western Australia in Canberra face the same problem. By contrast, no Member of Parliament in Bern is more than a three-to-four hour train ride or car drive fiom their electorate. Second, Swiss legislators sit shorter sessions

- four three-week sessions a year - and are not provided with the accoutrements of power their counterparts in Washington, Ottawa and Canberra receivee6The lobbyist in Bern wishing to speak with a Member of Parliament during the session cannot go to the legislator's office - MPs do not have offices. Instead, lobbying is truly done in the lobby behind the two legislative chambers, where tables and phone jacks are provided during legislative sessions.' Third, most committee hearings are held at times outside the sessions, and are distributed throughout the country. 124 Furthennore, federal legislators maintain a formal closeness to their cantons and their electorates through overlapping memberships. Data gathered by Steiner (1974: 126-

7) showed that more than 20 percent of the members of the lower house, the Nationalrat, were members of either a cantonal or communal executive. Official Swiss statistics indicate this percentage had decreased to 10.5 percent (2 1 of 200) by 1995.' Seven of the

46 members of the upper house, the Standerat, had memberships in cantonal govem~nents.~Such overlapping memberships are non-existent in Canada, Australia and the United States, especially among federal politicians. The time and travel requirements of those legislatures would make such overlapping memberships impossible.

The party system also reflects this decentralized federalism. The three traditionally largest parties (Free Democratic Party, Christian Democratic Party and

Social Democratic Party) have strength in virtually all areas of the country. This is somewhat misleading, however, because the parties are not truly national in nature, but are more accurately seen as the coming together of separate cantonal parties. Federal candidates are selected at the cantonal level, and it is not uncommon for the cantonal parties to bolt fi-om the position of the national party on issues put before the voters

(Steiner, 1974: 40-1; Kobach, 1994: 126-7, 197). As Ker. (1974:20) puts it, "the area of electoral competition is the canton, not the nation."

Direct democracy

The Swiss use direct democracy devices more extensively than any other people in the world. At the federal level, ballot measures are scheduled to be put to the voters every three months, although on some occasions the voting date is not actually used.

Ma.cantons use concurrent dates to hold their ballot measures. More than 400 issues have been put to a national vote.

The use of direct democracy in Switzerland, at least in the German-speaking areas, has a seven-century background. The Alpine cantons developed the institution of the Landsgemeinde, an annual open-air assembly of all male citizens that was the cantons' sovereign lawmaking body. However, some writers (Treschel & Kresi, 1996:

186) also note the importance of the French influence near the end of the 18th and beginning of the 19th century, pointing out the inclusion of mandatory constitutional referendums in the first Swiss constitution in 1798 emerged out the influence of the

French Revolution and the ideas of popular sovereignty. Such a basis for direct democracy has some common ground with the populist urge leading to direct democracy in the United States and Australia.

Ballot measures at the federal level are put to the people via four methods

(Kobach, 1993: 15):

Constitutional referendum - All constitutional amendments passed through the

federal legislature are put to popular vote.

Constitutional initiative - Voters may initiate popular votes on constitutional

changes by collecting 100,000 signatures. The federal legislature has the ability

to offer a counterproposal which would be placed on the same ballot.

Facilitative referendum - All laws or decrees of the federal legislature can be

challenged if 50,000 signatures or the votes of eight cantons are gathered within 90 days. The cantonal option has never been used. A successful direct vote

overturns the legislation.

Treaty referendum - Treaties may be brought to a vote if 50,000 signatures are * gathered. Treaties on membership in supranational organizations such as the

United Nations are considered constitutional amendments and automatically

brought to a vote.

The first three methods require a double majority for approval. Not only must the measure be approved by a majority of those casting votes, it must gain a majority in 12 of the 23 cantons, with the six half-cantons being measured as 0.5 for that purpose.

Treaties for membership in supranational organizations also require a double majority, but treaties brought to a vote may be rejected by only a popular majority.

The use of the referendum, both obligatory and facilitative, gives the people (and by virtue of the double majority, the cantons) sovereign power over legislation. The

Federal Council introduces most legislation, but has no signatory or veto power at the end of the legislative process as U.S. executives do, or, in a mostly formal way, as Australian or Canadian heads of state do. In Switzerland, that veto goes to the people. Declining to put forward a challenge during the 90-day period is an indication of sovereign assent to the bill (Kobach, 1993:41). Steinberg (1995:256) cites the pre-1790s constitution of

Canton Schwyz:

... the May Landsgemeinde is the greatest power and prince of the land and may without condition do and undo, and whoever denies this and asserts that the Landsgemeinde be not the greatest power nor the prince of the land and that it may not do and undo without condition is proscribed. Let a price of one hundred ducats be on his head.

The cantons and communes also make extensive use of direct democracy devices. Article

6 of the federal constitution requires the cantons put their own constitutional changes to the voters either through referendum or Landsgemeinde (Delley & Auer, 1986: 86). All cantons pemit citizens to bring forward initiatives, not only of the constitutional type as used at the federal level, but also for regular legislative items, and a slight majority have provision for some type of facilitative referendum.

All but eight of the cantons also have provisions which forbid the hding of projects over a certain cost without first referring them to the people in a referendum, a process similar to local bond issues in U.S. local government. For example, in

Neuchatel, capital expenditures of more than FS300,OOO must be approved in a referendum (Rohr, 1987:2 12).

The most striking feature of the Swiss use of direct democracy is that the initiative and referendum are seemingly majoritarian devices in a system that otherwise strives for consensual decisions. Switzerland is described in a wide range of literature as a consociational democracy (Steiner, 1974; Lijphart, 1979; Lehmbruch, 1993). The central idea of consociational democracy is conflict regulation through four means - proportional representation, grand coalitions, mutual vetoes and segmental autonomy (Lijphart, 1979:

25). The structure of the Federal Council, in particular the magic formula agreed upon by the major parties, as well as the provision of segmental autonomy through the power retained by the cantons and communes, demonstrates consociational practices are present in Swiss politics. This process of consensus through elite accommodation goes still

further. Not only do four parties share executive power, but a wide range of interest groups are brought into a formal consultative process. Art. 32 of the federal constitution

states "the appropriate economic groups are to be heard before the laws are made."

Four major groups, collectively referred to as the Verband and representing commerce and industry (Vorort des shweizerischer Handels- und Indsurienverein), small business (Schweizerischer Gerwerbverband), trade unions (Sch.,ve~erischer

Gerwerkschaftsbund) and farmers (Schweizerischer Bauemverband), were created late in the 19' century and early in the 20"' century, and each continues to receive financial support from the government (Kobach, 1993: 31). These groups, especially the first three, are consulted on all important issues, even those that do not directly affect their interests while smaller, more narrowly focused groups, are consulted on proposals more directly involving them (Kobach, 1993: 148).

Enhancing this consultation are the multiple roles of Swiss legislators. As mentioned above, some members of the federal legislature are also members of cantonal executives. Others hold leading positions in key interest groups. Such members understand their position in the legislature as representing not merely the interests of their canton or party, but of their interest group. Because serving in Parliament is a part-time occupation, one can work full-time as the leader of an interest group (Steiner, 1974: 122).

A dozen of the 200 members of the Nationalrat list lobbyist as their primary occ~pation.'~

Within this consensual system, direct democracy, and in particular the facilitative referendum, serves as a device which demands that the consensus reached is a broad- 129 based one. A consensus that leaves a major interest group out risks that group forcing a referendum on an issue. As Blankart (1993:9 1) writes:

The majority in Parliament can never be sure that an approved bill will become statutory law because opposing groups may launch a referendum and win the vote if the issue is contested. It would be unwise, therefore, for the parliamentary decisionmakers to govern with break-even majorities. Rather, they are well advised to base their decisions on broad majorities and to make compromises in order to encompass all possible opponents who could start a referendum. Broad- based majority decisions are generally less contested in the public and therefore have a higher chance to escape the verdict of the referendum than decisions by small majorities (though there is no guarantee for them).

In fact, the convention of sharing seats on the Federal Council emerged from the use of the referendum device by Catholics late in the 19"' century. The basic design of the current Swiss state was put in place in 1848. In 1874, a "total revision" of the constitution was approved which introduced the facilitative referendum. The Swiss federal government of the 19"' century was controlled by the loosely organized Radicals, victors over the conservative Catholics in the brief Sonderbund civil war of 1847 which led to the introduction of the 1848 constitution (Steinberg, 1996: 43-50). The Radicals held all seven seats on the Federal Council and controlled both legislative branches.

With the introduction of the facilitative referendum, Catholics had an avenue by which they could overtum the laws passed by the federal government. They brought 14 facilitative referendums forward in the next nine years, getting popular passage of 11 of them and overturning a wide range of proposals (Kobach, 1994: 27). In particular,

Catholics used the referendum to protect the powers of the cantons from encroachment by the federal government, thus enabling cantons with large Catholic majorities, such,as 130 Luzern, to maintain their autonomy. Among the federal laws overturned were measures on federally established voting rights, the creation of a federal education secretary, and the creation of a federal justice department (Kobach, 1994: 27,71). This pressure continued until 1891, when a prominent Catholic conservative, Joseph Zemp, was elected to the Federal Council and in an unusual move, was immediately made President

(Kobach, 1994: 28). The Catholic conservatives, who became formally organized as the

Christian Democrats in the late 1960s, gained a second seat in 1919, with the Fanner's party, now the Swiss People's Party, getting one in 1929, and the Social Democrats receiving one seat in 1943 and the second in 1959 to complete the 2:2:2: 1 magic formula

(Steiner, 1974: 33)."

Today, denying the considerations of a major political party or interest group can bring down a bill. As Kobach (1994: 134) points out, "The opposition of any single verband is usually decisive in bringing about the rejection of an initiative or law." In

1992, the smallest of the four "government" parties, the Swiss People's Party, and the main farmers' interest group, despite the support of the other major parties and groups, opposed the treaty referendum which would have brought Switzerland into the European

Economic Area as a first step toward joining the European Union (Jahrbuch: 376). It is also not uncommon for a party to split on a ballot measure. In particular, cantonal branches of a party may take a different line fiom the federal party, although this happens somewhat less often in the Social Democratic Party than in the other three. It should also be noted that on some occasions, the Swiss people reject the near-unanimous recommendations of their elites. They seem especially suspicious of pay increases or the 131 extension of other benefits to federal legislators. In 1962, a facilitative referendum was

organized by the tiny Free Citizen's Party, which had legislative representation only in the

canton of Aargau, where it had but five of the 200 seats. Even so, the referendum

succeeded in defeating a proposal supported by all four of the government parties to

increase legislative pay, and did so by a more than 2-to- 1 margin (Steiner, 1974: 19).

Initiatives may also serve as a consensus-building device. As pointed out in

chapter two, initiatives serve to put issues onto the political agenda that might not

otherwise be heard. But in many cases, once the initiative is put forward, the government

uses the same consensual practices to avoid a final conflict at the ballot box. In many

cases, agreement is reached with the sponsors of the initiative and it is withdrawn. This

can be done simply through negotiation, or through threatening to put a government-

written counter-proposal on the ballot. Because the percentage of successful initiatives is

quite low, there is incentive for groups to accept half a loaf. Statistics provided by

Kobach (1 993: 109-10) indicate that fkom 1891 to 1992, some 62 federal initiatives (a third of all those submitted) were withdrawn. In 54 of these cases, the initiators received some sort of concession fiom the government.

In some cases, the govemment may even react to an initiative which has gone to the voters and failed. After a 1989 initiative to abolish the army received a surprising 35 percent of the vote, the Federal Council moved quickly to introduce reforms to the citizen army, in which every Swiss male is obligated to serve in a status similar to that of reservists in North America. A 1991 constitutional referendum, which was approved by the voters, permitted the creation of alternative service programs for conscientious 132 objectors, who previously were forced to serve a jail tern. By 1995, a broad reworking

of the service system had been introduced which cut down the obligation fiom 30 years to

22 and reduced the minimum number of actual active service days in that span fiom 33 1

to slightly less than 300.12

To summarize, Swiss direct democracy operates in a system in which consensus- oriented politics are well established. The use of the facilitative referendum is largely to enable those who are left out of the consensus to take their case to the people. Because of this, the effort is made to go far beyond minimum winning coalitions. The initiative provides an alternative means, usually for those who are seeking to expand the national agenda. Furthermore, as discussed more klly below, ballot measures, in particular referendums, have served as a brake on the expansion of federal power. In regard to both consensual politics and federalism, direct democracy appears to work as an amplifier, expanding their importance and entrenching them within the system.

AUSTRALIA

In a strictly quantitative sense, the role of direct democracy in Australia has been more limited than in Switzerland or the American states in this study. Nonetheless, the constitutional requirement for popular approval of amendments to the federal constitution has played an important role in Australian political history and is instructive on a number of levels. First, it is worthwhile to briefly discuss the development of the constitutional referendum in Australia, and the role of populism in that development. Second, the referendum has served as one of the few brakes on Commonwealth power that has otherwise expanded substantially. Third, the Australian case is noteworthy for the

interplay of political parties and direct democracy.

Introduction of the constitutional referendum

The referendum in Australia has a history which slightly predates the

Commonwealth. Australia is unique among the four countries in that the initial

federation required direct popular approval in the constituent units. Although the initial process, which can be said to have begun with a convention in 1891, began more along

Canadian lines, with representatives sent fkom the various colonial legislatures, it became over the course of the decade one with more popular involvement.

By 1893, pressure was being applied by various interest groups, particularly fkom business interests in Victoria, for a new process. A Premiers' Conference in 1895 accepted the suggested new guidelines. Among these were the popular election of the convention members and rules requiring the final document to be refereed back to the people for referendums in each colony (Golan, 1955: 188). A convention based on these guidelines began in 1897 and produced a document sent to referendum in 1898. The initial document received approval in three colonies, was amended, and returned to the people again in 1899, when it received approval from all but Western Australia (Golan,

1955: 189). Western Australia joined only at the last moment, as its colonial government did not send the document to the people for approval until London was ready to give assent. 134 The design and result of the Constitutional Conventions were to some degree

influenced by their taking place after the development of the other three countries in this

study. The development of a federal state using Parliamentary government out of a

group of formerly separate colonies had precedent in Canada in 1867. Canada, however, left aside the issue how to amend the founding document until 1982. Hence the U.S. and

Swiss examples were more germane, and Galligan (1995: 115) points out both were considered as models for constitutional amendment. The 1891 conference focused more on the U.S. approach of using state legislatures to ratify constitutional amendments, but as the decade went forward, the Swiss practice of using referendums requiring a double majority gained support and was eventually selected (LaNauze, 1972: 286-7).

The introduction of the referendum may have some tenuous links to its introduction in some U.S.states at a similar time. In the early 1890s, literature on the use of direct democracy in Switzerland diffised worldwide (Scarrow, 1999: 276). It found a receptive audience among populists in the U.S., as was discussed in chapter two. Brugger and Jaensch (1985: 8-1 1) claim populism on both the right and the left was at an apogee in Australia during the 1890s because of an economic downturn which had its most severe effects in rural areas. This was the same worldwide depression which prompted the growth of populism in the United States. While the push for a more popularly based amending system does not appear to have come &om rural interests, one can reasonably speculate that populist growth put the ideas into play where they might not otherwise have been. Referendums and Commonwealth power

The majority of constitutional referendums in Australia have focused on two

areas. One, regarding changes in process (for example, alterations in the election of

Senators) has been reasonably successll. The other major area, the division of powers,

has witnessed less success. Some 24 of the 40 constitutional referendums in Australia

have sought to give the Commonwealth government expanded power in areas which the

constitution grants powers to the states or to both levels concurrently, but only three have

passed. Among the successful efforts were the granting to the Commonwealth the power

to control borrowing by the states through the Loans Council in 1928, the

Commonwealth gaining responsibility for social services in 1946, and the

Commonwealth gaining the power to pass laws regarding Aboriginals (1967).

Most of the efforts to expand Commonwealth power have occurred under Labor

Party governments. For many years, the Labor Party had as part of its platform a plank

which called for the creation of a unitary state. As late as 1979, a future Labor Prime

Minister, Bob Hawke, gave lectures calling for the abolition of the states (Galligan, 1995:

124). While no referendum ever was that far-reaching, Labor initiatives up through 1973

were of a centralizing nature. In particular, Labor sought to expand the Commonwealth's power over the economy.

Since the 1970s, however, the use of referendums as a method of providing wider power to the Commonwealth has waned. This has been due in large part to the success of two alternative methods, use of the courts and intergovernmental negotiations (Galligan,

1995: 131). As pointed out in the section on fiscal federalism, since 1920 the High Court 136 has been generally sympathetic to Commonwealth demands for wider powers, although

not to an unrestricted degree. The Commonwealth has also been able to extend its power

through intergovernmental relations because of its strength with respect to the spending

power. Furthermore, as Labor moderated and moved away fiom a focus on

nationalization toward policies more along the lines of welfare-state liberalism, its desire

for massive centralization of power declined. Hughes (1 994: 164) claims that, in addition

to broader reasons, the general populace has a suspicion of the federal government that

mitigates against the approval of centralizing proposals. He writes, "To the extent that

Canbema is perceived as a remote, hostile, and selfish force in Australian politics, any

proposal to tamper with the federal constitution is perceived as a potential Trojan horse

for

.. . the existing procedure of section 126 has an obvious bias in favor of pro- Commonwealth amendments being proposed, but these have to run the gauntlet of a thoroughly federal and popular referendum procedure before they can be ratified. Such disparity between the initiation and ratification parts of the Australian amending formula helps to account for the historic pattern of referendum results - so many proposals put to referendum but few being passed.

One might then consider the flow of power to the Commonwealth as a three-fork stream.

The referendum device can be seen as the stream which potentially carries the most volume, as it has fewer restraints than judicial interpretation or intergovernmental negotiation. The dam of popular skepticism, however, has blocked the referendum stream. The dam thus halts one stream, meaning the total volume of the flow of power to the Commonwealth is restricted, but not stopped. 137 Centralizing amendments are not the only ones which have had difficulty gaining

approval. Only five of the 16 remaining amendments have gained approval - a better rate,

to be sure, but hardly overwhelming support. A fear of centralization is not the only

factor preventing amendments. The nature of the party system also plays a role.

Political parties and the referendum

As a Parliamentary system, Australia has highly disciplined political parties.

Furthermore, these parties and the party system in which they operate are competitive

rather than consensual. The divide between Labor and the Liberal -National coalition is

deep. This party system has been identified by a wide range of writers as a major

impediment to the passage of constitutional referendums.

If both major party groups agree on an issue, the measure has some hope of

passage. However, it is not assured. The opposition may be successfblly led by a smaller

party, or by state branches of one of the major federal parties (Aitken, 1978: 133). When

the major parties disagree, such as was the case in the two referendums in 1999 on

becoming a republic and adding a preamble to the constitution, passage is nearly

impossible. Furthermore, after an early stage where the parties had some consensus on

constitutional issues (Aitken, 1978: 133), the likelihood of disagreement has been very high. Because of the competitive nature of party politics in Australia, the party out of power is quite likely to oppose almost anything put forward by the Government, simply to carry out their role as the Opposition and avoid giving the Government any opportunity to score political points. This has been especially true of the Liberal- National coalition. Labor has been somewhat more willing to give support to

referendums when in Opposition.

A notable example of the Opposition dissenting against proposed amendments

occurred in 1988. In 1985, the Labor government of Bob Hawke concluded the

Constitutional Convention that had met intermittently since 1973 had run its course, and

put together a Constitutional Commission that it was hoped "could skirt the partisanship

that had killed most initiatives for constitutional 'reform in the past" (Galligan, 1996:

124). The commission, which operated in some respects like the Spicer Committee on

Canadian constitutional reforn in the early 1990s, was given a vague fiame of reference,

and produced a broad set of proposals in 1988. Out of the proposals, the Hawke

government put four proposals before Australian voters in 1988:

To provide for four-year maximum terms for members of both Houses of the

Commonwealth Parliament.

To provide for fair and democratic parliamentary elections throughout Australia.I3

To recognize local government.

To extend the right of trial by jury; to extend fieedom of religion; and to ensure

fair terns for persons whose property is acquired by any government14

The proposals were not overly controversial. In fact, most elements of the Liberal Party, the senior member of the non-Labor coalition, would have been thought to be in favor of the second measure, and the Liberal Party platform contained a plank in favor of recognizing local government (Jaensch, 1994: 59). Nonetheless, the Liberal Party came out as opposing all four. Labor, on the other hand, in particular Attorney General Lionel 139 Bowen, having argued for a non-partisan process, gave only lukewarm support once it

proposed the measures. The result was that all four proposals were crushed. Not only did

they fail to pass; all four failed to pass in any single state (Hughes, 1994: 160).

The role of parties in Australia can be seen as an interesting contrast to that of the

parties in Switzerland. In each country, the conser.sus of the major parties gives a ballot measure the greatest chance of success. It is there, however, that the similarities end. In many, but certainly not all cases, constitutional referendums reaching the voter in

Switzerland will have the support of the four major parties. In Australia, the government can use a bare majority in Parliament to put a referendum onto the ballot. Consensus improves the possibility of success (for an example, note the referendums of 1977), but is not required, and given the competitiveness of the system, is unlikely.

AUSTRALIA AND SWITZERLAND COMPARED

A comparison of the use of direct democracy at the federal level in Australia and

Switzerland establishes a handful of conclusions. Foremost among these are the roles of political parties in direct democracy and the role of referendums in putting a break on centrifugal forces in a federal system.

It is apparent from the experience in both countries that all-party support for a ballot measure is an important factor in its success. Measures lacking the support of even one of the major parties have, at best, minimal chances of passage. The difference between the two systems is in the nature of the party systems. The Australian system is competitive; the Swiss system is consensual. Therefore, the likelihood of all-party agreement is much greater in Switzerland, as it occurs with much greater frequency

generally. Furthemore, even those outside the governing parties are consulted on issues.

This process, used for all legislation because of the facilitative referendum, is merely

extended to the constitutional referendum process. In Australia, the governing party,

assuming it has control of the Senate, needs only to maintain unity of the Parliamentary party to ensure passage of normal legislation. The need to consult or to negotiate with the opposition is not present. Such a process, however, is less well suited to success in a referendum campaign. Further, the opposition parties have little incentive to support the referendum, even if it is consistent with their principles.

The second point of comparison is the role of referendums in blocking expansion of federal power. This effect is apparent in both systems. Table 6.1 indicates the fates of centralizing proposals in the two countries. The difference, however, is in the alternative methods the federal level has used in gaining power. In Australia, the linked variables of favorable judicial interpretation and fiscal dominance have enabled the Commonwealth to put the states in a truly subordinate position. It should be noted that one portion of this fiscal dominance, the creation of the Loans Council and the federal control of state debts, did emerge out of a successful referendum in 1927. Nonetheless, most expansion of federal power has occurred despite the presence of the referendum, rather than because of it. In Switzerland, the federal government has had neither of these advantages. The federal government has never had a fiscally dominant position. Indeed, the presence of the referendum has seen to that. 141 CANADA

As has been stated previously, the use of direct democracy in Canada has been

quite limited, especially at the federal level. But the last two federal referendums held in

the Canada - on the Liberal government's commitment not to introduce conscription in

1942 and the vote on the Charlottetown Accord in 1992 - do provide some useful

contrasts and comparisons to the processes used in Australia and Switzerland. This

section will briefly consider the 1942 referendum, more fully discuss the 1992

referendum on the Charlottetown Accord, and briefly note the sovereignty referendums in

Quebec in 1980 and 1995.

A small proviso should be noted here. Canadian referendums to the present day

have technically been plebiscites in that they are not legally binding. While political

commitments may make the plebiscites binding in a practical sense, legislation must still

be introduced and passed through "normal" channels before the will of the people can be

implemented. A possible explanation for this may be rooted in the role of the Crown in

Canadian politics, and more specifically by the interpretation of that role by the Judicial

Committee of the Privy Council.

In 1916, the Manitoba legislature passed the Initiative and Referendum Act,

which enabled indirect initiatives to be introduced through the signatures of eight percent

of the electorate of the previous election, and facilitative referendums to be used through

the signatures of five percent of the electorate (Boyer, 1992a: 88-9). But the bill was struck down on the grounds it placed limits and requirements on the Lieutenant-

Governor. This legislation was introduced in an era when nearby states to the south of 142 Manitoba were implementing the devices of direct democracy. There, no constitutional

limitations were found. In Canada, however, the prerogatives of the Crown inhibited the

possibilities of binding direct legislation, although not, as Smith (1995: 70-1) points out,

plebiscites. W.L.Morton (1957: 349-SO), who concedes the legislation may have been in

keeping with the tenor of the times, nonetheless states its passage by the Manitoba

legislature "suggests either an imperfect grasp of the principles of parliamentary

government, or considerable political naivete."

It is unclear, and perhaps even unlikely, whether such a ruling would be repeated

today. To this point, legislation on binding referendums for constitutional approval in

British Columbia and Alberta, and initiatives in British Columbia and Saskatchewan,

have not been challenged. A Canadian Supreme Court in the era of the Charter of Rights

and Freedoms would be a far different setting than the JCPC. The idea the current court

would fall back on the prerogatives of the Crown to strike those measures down seems,

prima facie, to be far-fetched.

The 1942 Conscription plebiscite

In 1939, Canada declared war on Gemany. Mindful of the divisiveness of conscription during World War I, the government of Mackenzie King won reelection in

1940 on a platform of "no conscription for overseas service" (Boyer, 1992a: 36). By late

1941, the government was divided on the issue and being pressured by the pro- conscription policies of the opposition Conservatives (Johnston, et al, 1996: 256).

Seeking a way out and having rejected both calling a general election to seek a mandate 143 and a true referendum on the issue, King instead sought a plebiscite, not directly on conscription itself, but rather on whether the government should be fieed fiom its commitment on the issue. The precise wording of the ballot measure asked "Are you in favor of releasing the government fiom any obligation arising out of any past commitments restricting the methods of raising men for military service?" (Boyer, 1992a:

39).

The plebiscite was held on April 27,1942. It received massive support in

Anglophone areas, and thus passed with nearly 63 percent support nationally. The seven overwhelmingly Anglophone provinces all provided support levels over 70 percent, and in New Brunswick, 69.2 percent approved the plebiscite. The plebiscite failed to receive support in Quebec, however, and in the Francophone community more broadly, failing to garner a majority in any federal riding with a Francophone majority (Boyer, 1992a: 42).

Johnston, et a1 (1996: 257) argue the primary aim of the plebiscite was to get the

Liberal Party out a tight spot, demonstrating to its Quebec members that support for conscription was strong among Anglophones, and equally to its other members that opposition was strong in Quebec. Further, it undercut the Conservatives by taking their main issue away, without actually giving a commitment to change policy, as King delayed taking real action on conscription until 1944.

The conscription debate, however, raises a more fundamental point. The plebiscite passed nationally, but King's no-conscription pledge had been directed primarily at Francophones. King interpreted the result as letting him out of the platform commitment, but who were "the people" King made the commitment to? Were they the Canadian people? Or were they some subset of the Canadian people, either

Francophones, or, to define the subset in federal terns, the people of Quebec? Double majorities in a federal system can allow both understandings of the "people" to have a voice. In this particular case, no formal mechanism for recognizing double majorities existed. Further, short of a veto for Quebec, any double majority would have provided the same result. Informally, King responded to the outcome in a Janus-faced manner, treating the national result as though he had been fkeed fiom his non-conscription commitment, but, recognizing the divisions present, not acting on the latitude granted him for another two years.

Finally, the 1942 conscription referendum raises an additional series of questions.

The measure on the ballot was not a binding referendum put before the people through established rules and procedures, but rather an ad-hoc, non-binding plebiscite. As such,

King was able to interpret the results however he saw fit. How do true referendums with binding results and established rules differ? Is this a question that would have been brought to a referendum, and if so, would it require a double majority for passage? As a defense and military issue, it would seem to be one with a strictly national jurisdiction; thus, save for some constitutional question being involved, a double majority would not be necessary. For the interests of Francophones, or of any particular province, to be included in the final decision despite their minority status, the system would have to be consociational in some sense, granting a veto to the opposed group. Charlottetown Accord

The "referendum" on the Charlottetown Accord provides an excellent case study which can be used to examine some of the assumptions taken from the Australian and

Swiss use of ballot measures. Before examining these assumptions, I shall provide an overview of the events leading to the referendum. From there, a handful of matters shall be tackled - the role of parties, the tendency of referendums to protect a dispersion of power, the role of double majorities, and the danger of multi-layered referendum questions. The Charlottetown Accord can be seen as the climax of nearly three decades of megaconstitutional politics in Canada (Russell, 1993). Two events in that history are crucial - the implementation of the Constitution Act, 1982, and the failure of the Meech

Lake Accord in 1990.

The implementation of the Constitution Act, 1982, is notable in this context for two reasons. First is the rejection of the agreement by the government and legislature of

Quebec. This set the agenda for the next decade, especially for the Meech Lake Accord, but also for the Charlottetown Accord. Second, the Constitution Act, 1982, introduced a

Charter of Rights and Freedoms. As Cairns (1992) states:

...the Charter enters the constitutional reform process in that, by connecting the citizenry directly to the Constitution and by linking particular categories of Canadians to specific constitutional clauses that they view as theirs, the Charter transmits the message that some degree of citizen participation in the constitutional reform process is logically necessary. Further, the fact that "rights" is the medium for this constitutional connection is psychologically empowering in a way that a more instrumental connection would not be. 146 The Charter was placed into the Constitution Act, 1982 in part because it fit the Trudeau

vision of a rights-based liberal democracy, but also because of pressures fiom various

interest groups, in particular women's groups. Such groups could be expected to

jealously guard their status and their place at the table, indirect though that status may

have been (Russell, 1993: 114-5). Therefore, the negotiations for the Meech Lake

Accord began with a new burden which may not have been immediately recognized. Past

federal-provincial negotiations could be held in some respects outside the public glare. A

deal might be unpopular, but would not be seen, except perhaps in Quebec, as

illegitimate. In the new era, the closed process was seen by some as illegitimate, because

it was carried out by "1 1 white males" behind closed doors (Gibbins, 1990: 262). With

the failure of Meech Lake Accord, cries were heard that never again could the

. constitution be discussed in so closed a manner.

As a precursor to new negotiations, a wide-ranging consultative process took

place that included "citizen's forums," travelling Parliamentary committees and other

public discussions. This process culminated in the Beaudoin-Dobbie Commission, which

in February of 1992 produced a 125-page report, A Renewed Canada, that became the

skeleton for new negotiations (Russell, 1993: Chapter 10). Within six months, the

federal and provincial governments had negotiated, again primarily behind closed doors,

a wide-ranging agreement that had, as its central tenets, the following items:

a A Canada Clause stating fundamental values, including the right of Aboriginals to

"promote their languages, cultures and traditions," the recognition that Quebec is

a distinct society, and sexual equality. 147 a The creation of an economic and social union the provisions of which were to be

made specific at a later date, but were to include guarantees of universal health

care, equalization payments and fkeer interprovincial trade.

a Reform of national institutions. The Senate would be made smaller, but made an

elected chamber, with a full veto over natural resource legislation, and the ability

in some other cases to force a vote of a combined sitting of the House of

Commons and the Senate. In exchange, the House of Commons would add seats

for Ontario, Quebec, Alberta and British Columbia to more accurately create "a

rep by pop" chamber. Somewhat paradoxically, a floor would be created for

Quebec's representation, in that Quebec would be indehitely guaranteed a

minimum of 25 percent of House of Commons seats regardless of its proportion

of the Canadian population.

a The recognition of Aboriginal people having an "inherent right to self-

govemment."

a Alterations to the amending formula, which would broaden the range of items

needing unanimous support of the provinces for passage.

a An alteration of the division of powers which would provide explicit provincial

jurisdiction in a wide range of areas, and would also allow provinces to opt out 06

new federal programs with full compensation.

Some of the impetus for the Charlottetown Accord had been the threat of a Quebec referendum on sovereignty if no new offer were forthcoming after the failure of Meech

Lake. The Quebec government had committed itself to such a referendum in either June 148 or October of 1992, leaving itself room to make the vote either one on sovereignty or a

pending offer (Russell, 1993: 162-3). The intergovernmental negotiations of the summer

of 1992 provided that offer. Quebec, however, was not the only province committed to a

referendum. In the wake of Meech Lake and the concerns about a process that was too

closed, Alberta and British Columbia had passed legislation committing each province to

a referendum on any constitutional agreement. With three of the four most populous

provinces committed to sending the accord to the people, the process was broadened so

that the federal government would hold a nationwide vote, although allowing Quebec to

maintain its own legislated mechanisms for the referendum.

On October 26, 1992, Canadians went to the polls to vote on the following

question: "Do you agree that the Constitution of Canada should be renewed on the basis

of the agreement reached on August 28,1992?" As the accord would need to gain passage in all 10 provinces to be ratified, a unique expansion of the double majority

concept would be required for passage. Passage would not need merely a majority of the people as well as majorities in a majority of the 10 provinces, but rather would require the much higher standard of passage by a majority of votes in each and every province."

In the event, the Charlottetown Accord was decisively rejected. It gained approval in only four provinces, and in one of those (Ontario) by the narrowest of margins. Six provinces, including Quebec, rejected the accord, and overall, 55.0 percent of those casting votes in a 74.7 percent tumout voted no. What, then, does the

Charlottetown Accord say about the trends evidenced in Australia and Switzerland? Role of the parties

In examining Australia and Switzerland, it has been noted that all-party support

for a referendum provides it with a much greater chance of success. On this basis, the

Charlottetown Accord should have been in good shape. At the federal level, the three

major parties, which had combined to receive more than 95 percent of the vote in the

1988 federal election, supported the agreement. Only the sovereigntist Bloc Quebecois,

which had broken away fkom Progressive Conservative caucus, and the Reform Party,

which held just one seat gained via a by-election, were opposed. This support extended

down to the provincial level. Not only were the 10 governing parties supporting the

referendum, seven of the 10 Official Oppositions also supported it. Again, Quebec

sovereigntists, this time the Parti Quebecois, were opposed. But the B.C. Liberal Party

was split, with its leader, Gordon Wilson, supporting the "no" side and elements of his

caucus supporting "yes." A similar situation existed in Manitoba (Russell, 1993: 221).

A closer look reveals that all-party support may have had an impact. Table 6.2

shows the three provinces where the Official Opposition was either opposed or split on

the referendum were three of the four provinces with the highest percentage of "no" voters. Only Alberta breaks the full correlation, and in Alberta, the opposition of the

Reform Party may have played a role. Reform was already well established in the province, running at 35-40 percent in public opinion poll^'^, and its lone federal Member of Parliament was fiom an Alberta riding. The presence of elected politicians in opposition to the accord probably did have an effect. TABLE 6.2

CHARLOTTETOWN ACCORD REFERENDUM RESULTS WITH POSITIONS OF OPPOSITION PARTIES

PROVINCE "YES"VOTE OPPOSITION POSITION

PEI YES NFLD YES NB YES ONT YES NS YES SASK YES PQ NO ALB YES MAN NO B.C. NO

SOURCES: Vote totals - Chief Electoral Officer of Canada. Opposition positions - Peter Russell, Constitutional Odyssey. Decentralization

It was argued above that in Switzerland and Australia, referendums have acted as

a brake or dam on centralization. Yet Charlottetown, in most regards a decentralizing

document, failed. Three possible reasons for this can be suggested.

First, while the experience in Switzerland and Australia shows that measures

attempting to expand central power fare poorly, there is little information to support the reflexive corollary, that decentralizing measures do well. No such measures have been sent to the voters in Australia, and only a handfbl in Switzerland. Second, in the afterglow of the introduction of the Charter of Rights and Freedoms, many groups were opposed to a reorganization of powers away fiom the federal government. Charter groups feared a weakening of rights would accompany decentralization, and in the

Maison du Egg Roll speech, Pieme Trudeau mentioned the inability of poorer provinces to maintain social programs as one of a number of criticisms (Johnston et al, 1996: 69).

Third, while the deal was decentralizing fiom an English Canadian perspective, that was less clear in Quebec. From the Quebec perspective, the province was coming into a new agreement that would mean giving up a certain latitude and flexibility in negotiations.

For Quebec, the decentralizing measures granted to it were not enough to justify signing the Accord.

Even with these theoretically plausible arguments, at least one account (LeDuc &

Pammett, 1995) says views about decentralization had little to with the final outcome.

While those opposing greater powers to the provinces were more likely to vote no, their 152 numbers were relatively small (28 percent) and they had no statistically significant effect

on either the yes or no vote.

Com~lexity

A wide range of research in the U.S. has established that complex ballot measures

which seek approval for a group of items rather than a single item have less success

(Magleby, 1984: 142; Bowler and Donovan, 1998: 109). This has also held true in

Australia (Jaensch, 1994: 56). The Charlottetown Accord may well have been the

shining example of this. The accord consisted of 60 items, some of which had not been

completely fleshed out and would require fiuther negotiation after passage.

One may explain this through the analogy of a group of people ordering a pizza.

The group may be able to agree right away that it wants a pizza. It may even be able to agree on a topping or two, particularly if the toppings are bland, routine choices. But the more toppings that are added, the more chance there is of one person finding one particular topping so abhorrent that he or she chooses to have no pizza at all. This can also occur because someone insists that they will not eat the pizza without anchovies.

This may cause many members of the group to withdraw. In a referendum, putting a long laundry list of items under the umbrella of a single measure has the same effect. No matter how much they may like the back bacon and pineapple, they refuse to order a pizza with anchovies. In the case of Charlottetown, the Accord first had to overcome the pizza problem in building a deal among the premiers, then the people. It passed the first hurdle, but could not overcome the second. 153 Evidence of this comes fiom the Canadian National Election Study (Johnston et

al, 1996: 93-95). The research team narrowed the major items of the Charlottetown

Accord down to three for those outside Quebec - Senate reform, distinct society and

Aboriginal self-government. Those who supported all three elements overwhelmingly

voted yes. Those who supported two of the three were split. Those opposed to either

Senate refom or Aboriginal self-government but supporting the other two elements were

slightly more likely to vote yes (roughly 56 percent did so). Those opposed to distinct

society were slightly more likely to vote no (roughly 42 percent voted yes). Distinct

society appears to have been the anchovy in the pizza.

The Charlottetown process has much in common with the typical Swiss procedure

for not just constitutional amendment, but also for the passage of regular legislation. As

Russell (1993 : 191-2) points out, the Charlottetown Accord consultation process was

shaped like an hourglass. The process was wide-ranging in its earliest stages, narrowed down to the committees and commissions attempting to provide proposals, then narrowed further to closed-door meetings of the First Ministers. After the First Ministers reached an agreement, the process widened out again, as the people were asked to approve the

$ accord. The difference was that at no point until the First Ministers' negotiations was everyone at the table. Quebec was on its own set of tracks, developing a new list of proposals. The Charlottetown Accord, then, was not truly a consensus document that emerged out of broad public consultation, but rather one that emerged out of the negotiation between two differing consensuses. In Swiss negotiations, everyone is at the table fiom the beginning. UNITED STATES

In the case of the United States, the focus will be narrowed to a small set of questions. In particular, two issues will be delved into. First is the impact of referendums on minorities within the context of the American political systems. The second issue is whether or not ballot measures, in particular, initiatives, serve as a means for bringing new issues onto the political scene. To be more precise, do issues raised in state initiative campaigns find voice at the federal level?

The development of direct democracy in the U.S. has a varied history. Spatially, it has been much more prevalent as one moves fiuther west. Organizationally, it becomes more common as one moves down the jurisdictional ladder. No direct democracy exists at the federal level. It has some presence in virtually every state government, but as stated above, markedly more in the West. Direct democracy is most prevalent at the local level, with capital projects and in some cases regular school funding requiring the approval of voters in many jurisdictions. In this respect, direct democracy in the Swiss cantons may have more in common with U.S. counties and cities than with states.

Minorities

In chapter five, it was shown that in quantity, the use of initiatives and referendums against minorities has been quite limited. Nonetheless, a more detailed look at the relationship of direct democracy and minority rights is in order. The United States is the proper setting for this deeper examination because unlike in Canada and 155 Switzerland, the two numerically most significant minorities lack an important sub-

national jurisdiction or group of jurisdictions in which they are the majority. The black

community is widely dispersed throughout the United States. While a majority in some cities and counties, as well as the District of Columbia, the largest percentage of fican-

Americans in any state, according to 1997 figures (U.S. Census Bureau, 1998: 36) is 36.3 percent in Mississippi. The Latino population in the United States, while somewhat less broadly dispersed, similarly has no state where it is the majority, corning the closest in

New Mexico."

When initiatives described as "anti-minorityyyreach the ballot, they often receive a great deal of media attention. To critics of direct democracy, these issues provide a demonstration of the danger posed by direct democracy as a weapon for tyranny of the majority. Recent anti-ahative action initiatives seeking to eliminate racial considerations in university admissions, government contracting and other areas have occurred after the scope of the data, but other initiatives do fit within this context." Most notable is a 1994 initiative in California, Proposition 187, dubbed the "Illegal

Immigrationyyinitiative, as it barred the provision of various public services, including education, from being provided to illegal immigrants. It will be used here as a case study to raise a series of points about the impact of direct democracy on minorities.

While Whites make up only a little more than half of the California population, they turn out in much higher numbers than other groups and thus comprised 81 percent of the state electorate in 1994 (Tolbert & Hero, 1998: 226). According to data compiled by

Tolbert and Hero (1998:296), Proposition 187 passed with 59 percent of the vote, largely 156 on the basis of overwhelming White support. It was narrowly rejected by Blacks and

Asians, and comprehensively defeated by Hispanics, among whom the greatest impact

from the measure would have been felt.

On the surface, Proposition 187 might seem to be a prime example of minority

rights being threatened at the ballot box, particularly in light of the ethnic breakdown of

voting. However, this would be an oversimplification. Proposition 187 was never hlly

enforced. Its implementation was halted by a series of injunctions, and the measure was

ruled unconstitutional by a federal district court in late 1997. Although an appeal was

launched, it was dropped in July 1999, when Gov. Gray Davis, who had opposed

Proposition 187, signed an out-of-court agreement with opponents of the initiative, who

agreed to drop the appeal, effectively voiding most provisions of the measure

(McDonnell, 1999).

The other point that must be emphasized here is that provisions such as

Proposition 187 can just as easily come from legislatures. Latinos have tended to have

fewer members of the California House of Assembly than their proportion of the

population would indicate. Although 26 percent of the population in 1990 (Tolbert &

Hero, 1998:212), Latinos sat in just four of the 80 seats in California's lower house after the 1990 elections (Bowler & Donovan, 1998: 133), a number which remained the same

following the 1998 elections (Bustillio, 2000). Perhaps the most telling indicator is that many of the limitations placed on illegal immigrants were passed the following year by the federal government through normal legislative processes. If Proposition 187 was, as its critics suggested, discriminatory, it still does not indicate direct democracy is any

worse for minorities than representative democracy.

To summarize, Proposition 187 does raise the specter of a tyranny of the majority,

as do other such initiatives, but this is tempered by three points. First, as noted earlier,

while such initiatives do receive a great deal of attention, they are relatively uncommon

in comparison to matters regarding taxation or government structure. Second, the

protections provided by constitutional rights provisions mean something which is truly

discriminatory is likely to be struck down by the courts. Finally, it cannot be established

that direct democracy is more likely to be discriminatory than representative bodies. In

fact Proposition 187 seems, indirectly at least, to argue against that.

The upward movement of issues

One common claim of direct democracy advocates is that direct democracy, in

particular the initiative, allows issues to be brought into political discussion that elected

politicians might otherwise ignore (Cronin, 1989: 11). The basis of this claim is that

legislators avoid certain issues either because they fear that the mere discussion of some

issues (abortion might be one) will inevitably alienate one large bloc of voters or another,

or, more cynically, because financially well-off interests backing the legislator do not want the issue raised. The latter view, although not raised in the context of direct democracy, was held by Schattschneider (1960), who claimed business and financial interests served as gatekeepers holding certain issues off the political agenda. While this is only indirectly applicable to this study, it does shed some light on the co-existence of populism and federalism. In this instance, the use of direct democracy in federally

defined regions may provide an alternative means of putting issues on the national

agenda.

If direct democracy does serve as a way of allowing broader participation in

agenda setting, the question becomes just how far this influence goes. If popular support

through the initiative process gets issues onto state ballots and successfully approves

those measures, can that support then also translate into effective pressure for legislators

at the federal level to act on the same issues? If issues gain acceptance in a number of

states, then we might expect these issues to be addressed at the federal level. Such an

occurrence would also lend some support to the understanding of sub-national

jurisdictions as "policy labs" for testing federal policies. Two issues have been selected

for examination. The first is the tax "revolt" of the late 1970s. The second is the term

limits movement of the 1990s.

The most notable example of the initiative being used as part of a drive to lower

taxes in the late 1970s was California's Proposition 13 in June 1978. The Jarvis-Gann

Property Tax initiative sought to limit increases in both property tax rates and property

valuations. It passed handily, getting 65 percent support. Almost immediately, efforts to

put similar proposals on the ballot emerged in other states, some of which reached the

voters in general elections that fall. Cronin (1989: 199) writes "Proposition 13 served as

a model or inspiration for dozens of tax-cut measures in about half the states."

The five states focused on in this study all had tax-reduction initiatives put before them in either 1978 or 1980. Oregon rejected initiatives similar to Proposition 13 in both 159 years. The other three states passed somewhat different tax limitations, Washington

doing so in 1978, North Dakota and Montana following suit in 1980 (Citrin, 1984: 17).

To what extent, then, did the tax-reduction pressure brought to the fore by

Proposition 13 and other such measures reach the federal level? We can see evidence that it did in two aspects. First, it had an immediate impact even in 1978. Proposition 13 was approved by California voters in June as part of a primary election. In November, both the House of Representatives and the Senate experienced a swing toward the

Republicans, one news analysts of the time attributed largely to tax issues.lg The larger impact, however, and one which can be more readily established as growing out of the

1978 tax revolt, was on the 1980 elections. Ronald Reagan's Presidential campaign was based on a domestic policy of lower taxes and smaller government. In Thomas and Mary

Edsall's Chain Reaction, an account of the breaking apart of the Democratic Party's New

Deal coalition, they write,

The tax revolt was a major turning point in American politics. It provided new muscle and new logic to the formation of a conservative coalition opposed to the liberal welfare state.. . The tax revolt provided conservatism with a powefil internal coherence, shaping an anti-government ethic, and hlyestablishing new grounds for the disaffection of white working- and middle-class voters fiom their traditional Democratic roots (Edsall& Edsall, 1991 : 13 1).

The tax revolt is the most notable example of issues starting out as state-level initiatives and broadening into national issues at the federal level. Furthermore, the tax revolt gained its initial impetus almost exclusively through direct democracy.

Legislatures were generally still following a path of greater government action, a deeper 160 welfare net and relatively high taxes. The tax revolt initiatives set in motion changes which were both wide ranging and long term.

Other issues have moved fiom the state level to the national level but failed to find success in Washington, D.C. Throughout the early 1990s, term limits were a popular initiative item in the states. Term limits measures passed in 23 states. The issue found little success at the federal level, however. The U.S.Supreme Court struck down state-imposed limits on federal representatives as unconstitutional (U.S. vs. Thornton,

1995), and while the matter did reach the floor of the House of Representatives as a

Constitutional amendment, it failed to pass. Even candidates who strongly supported the concept eventually backed away fiom it. Perhaps the most notable example was George

Nethercutt, a Republican fiom Washington who defeated Speaker of the House Tom

Foley largely on the term-limits issue in 1994. Nethercutt claimed he would serve six years and leave, but later renounced that claim and was successful in winning a fourth two-year term in November of 2000.

Term limits was an issue which found voice through direct democracy. For obvious reasons, most legislatures are not interested in the idea. But while it found national voice, it never had national success, and is much less prominent at the end of the decade than it was at the beginning. Other issues which had some success in state and local initiatives found nearly no voice at the federal level. Some jurisdictions declared themselves nuclear-fiee areas, but the nuclear fieeze associated with such action was never seriously considered at the federal level, although as a defense issue it was a matter of solely federal jurisdiction. 161 From this brief overview, we can surmise that while success is not assured, the

initiative can be used to put issues on the agenda not only of the jurisdiction in which it is

used, but also to extend the issue to other jurisdictions, including the federal level. This

would be consistent with the policy-lab thesis. It also demonstrates that issues which

start out as popular initiatives can eventually be taken on by legislatures and legislative

candidates.

CONCLUSIONS

Some direct comparison has already been made within this chapter. Nonetheless,

some additional tying together should be undertaken here. In particular, it is worth noting

the differing evolutions of direct democracy devices within each state. In Switzerland,

the idea of direct democracy predates the creation of modem federal state, whether that be

in 1798, or more likely, 1848. The expansion of direct democracy at the federal level

through the last half of the 19' century can be seen as a natural, indigenous event. The

introduction of the devices into the United States and Australia owes at least something to

the pre-existing Swiss experience. In each case, however, the adoption of direct

democracy was not as complete as in Switzerland. In the United States, the states have

used the devices, especially in the west. In fact, the data in this study shows states such -

as California and Oregon use the initiative and referendum more frequently than the

Swiss cantons. But east of the Mississippi, the devices are rare, save for referendums on amendments to state constitutions. At the federal level, direct democracy is non-existent.

In Australia, only the referendum device is present. As a binding device, it applies only 162 to constitutional amendments. One can only speculate on the possible reasons for this, but some plausible hypotheses can be suggested. First, the lack of U.S. federal direct democracy can be understood in two ways. One, even now barely half the states use direct democracy beyond constitutional amendment referendums. The support for the devices may not ever have been broad enough to introduce federally, especially as it would require a constitutional amendment. Two, the structure of the federal government and the emphasis in the political culture and in schools on the separation of powers and of

"checks and balances" as protection fiom pure democracy may further weaken desire for federal direct democracy. Local resentment in the west may have been enough to change state-level attitudes, but was not enough for a national change.

Explaining the Australian case may be helped through reference to Canada. The

Canadian west experienced some of the same conditions that led to the introduction of direct democracy in the U.S. west, but the only true effort at the introduction of initiative and referendum was struck down as interfering with the powers of the Crown. It may be that responsible govemmeat works against the development of direct democracy, especially the initiative. Thus, while the populist urge led Australians to include popular consent within the amending formula, pre-existing notions of responsible government meant the idea of initiatives remained beyond the pale.*'

' The five cantons with only one seat in the Nationalrat elect their representative through first-past-the-post methods.

The party names, following the most common practice, are English translations of the German names. It is worth noting, however, that for three'ofthe four parties, their French names are somewhat different. The Free Democrats become the Radical Democrats, the Social Democrats become the Socialist Party, and the Swiss People's Party becomes the Union of the Democratic Center.

The Swiss People's Party was the biggest vote-getter in the Nationalrat elections, receiving 22.56 percent of the vote to 22.48 for the Social Democrats, but the Social Democrats gained 5 1 seats to 44 for the Swiss People's Party. The Nationalrat electoral system is widely descnied as proportional, but this is mitigated by rule that each canton is its own district. Zurich, which has 34 seats, and Bern, which has 27, provide highly proportionate representation with low thresholds needed to gain seats. This is not the case for five cantons having either two or three seats, and five cantons having just one seat are in fact first-past-the-post.

Based on results in the 1995 Nationalrat elections, four cantons are one-party dominant - Uri for the Free Democrats, Obwalden, Niwalden and Appenzel Inter-Rhoden for the Christian Democrats. Steiner (1973), however, identifies only Appenzel Inter-Rhoden as such. All four are Alpine cantons. Uri, the largest of the four, has a population of about 40,000.

'The federal government has never levied income taxes on a permanent basis, something Linder (1994:104) says is unique among the world's states.

In the 11-year period 1988-98, the Nationalrat was in session an average of 322.5 hours a year. Assuming the fairly leisurely pace of seven hours a day, Monday through Friday, this would be an average of 46 days a year, leaving another 14 weekdays free during sessions. The Standerat meets even less - an average of 174.6 hours a year. Information taken from http://www.parlament.~h~E/Statis~auer-Sessionen-e.h~.

'Information gathered on tour of Swiss federal legislature in May, 1996. ' See http://www.parlament.ch/E/StatisMe~fl-Zusammensetzung~NR~e.htm.

See http:llwww.parlament.ch/E/StatisMe~fl-Zusammensetzung~SR~e.htm. lo See http://www.parlament.ch~E/Stati~Merufl-Z~~ammensetzung~NR~e~htm.

Steiner (1973: 74-5) adds thatin the two respects the grand coalition model was not something new. First, many cantons were already using the practice in their exectutive. Second, compromise was required to get almost anything passed through the pre-1848 Diet, which required a bill to receive the approval of all cantonal delegations for passage. l2 A more detailed account of the changes to the army can be found in Chapter 7 of Johnathan Steinberg's Khy Switzerland, 2nd ed. (London: Cambridge, 1996). The primary goal of this proposal was to do away with the overrepresentation of nual districts and to ensure "one vote, one value." l4 Essentially, this proposal expanded to the states rules already in force for the Commonwealth. l5 Some opponents of the accord claimed that if nine provinces voted favorably on the referendum, the legislature in the tenth would be pressured to follow suit despite popular rejection. This view held particular sway in British Columbia.

'"ata courtesy of Angus Reid Group.

"New Mexico does not have a provision for initiatives, although it does have the facilitative referendum. The focus here will be on issues involving ethnic, racial and religious minorities. This limitation is based on the idea that "people" is a cultural term and such groups would be most likely to suffer exclusion from the "eople."

For example, see "Got Your Message," Time, Nov. 20, 1978: 22-24.

20 Western Australia did consider initiative legislation at about the same time as Manitoba. It passed through the lower house of the legislature, but was killed by the Legislative Council, the upper house (Hughes, 1994). CHAPTER 7 - CONCLUSION

The relationship between populism, as operationalized through direct democracy, and federalism is a complex, multi-faceted one. Through the last three chapters, the details of that relationship have been examined, first through a broad comparative lens, then through an examination of specific institutions and referendums within each country.

Having completed the survey, we are now prepared to return to the broad questions raised in chapter one and attempt to summarize the evidence presented.

THE QUESTIONS RAISED:

I) What is the relationship of direct democracy to the operation of federalism?

This question has focused on two areas - the impact of direct democracy on both the concentration of jurisdiction at the federal level and the negotiating processes of federalism. While the evidence is not totally conclusive, it can be broadly concluded that direct democracy does slow the flow of power to the center and also has an impact on negotiation.

In regards to the movement of jurisdiction, direct democracy has generally had the effect of slowing down the movement of power to the center. In particular, in cases where the federal government sends proposals to increase its power to the people, the chances of success are quite slim. This seems especially clear in both Switzerland and

Australia, where the largest number of cases are available. In each country, centralizing 166 referendums are approved at a rate lower than referendums as'a whole. This is not to say

that increased centralization has not occurred. In Australia, especially, the central

govemment has found other means of expanding its jurisdiction. Even so, direct

democracy has, to return to the metaphor used earlier in the study, been a dam blocking at

least one flow of power to the center.

The effect on negotiation is interesting. The Swiss experience seems to indicate

direct democracy may enhance consensus, forcing all viewpoints to be heard. First the

referendum and then the initiative worked in favor of expanding the participants in

consensual decision-making. Furthermore, this works both at the level of normal

legislation and constitutional amendments, in part because most of the amendments are

not of a mega-constitutional nature as the Charlottetown Accord was in Canada.

2) What is the impact on minorities, in particular minorities thatfind a voice through federalism?

The danger of direct democracy to minorities within a given jurisdiction appears

to be overstated, although it does exist. Ballot measures threatening or weakening

minority rights have been infrequent and generally unsuccessful in the jurisdictions

studied here. Setting aside for the moment the possibility of judicial review, one might

argue that any evidence of votes having restricted rights is evidence of some danger to minorities. The problem with this is two-fold - it makes no mention of possible abuses by legislatures, and, more importantly for this study, it ignores cases in which ballot measures have served to expand rights, in particular voting rights. More measures expanding rights have been successful than have those restricting them. 167 Minorities that are represented through sub-national jurisdictions fare quite well.

The double majority provision, as used in Australia and Switzerland, provides a means through which voices that would otherwise be subsumed within the greater society can be identified and heard. Individual citizens are thus heard in two contexts - the national context and as a member of a federally defined community.

3) What are the intervening variables that impact the relationship?

This question focused on two sets of variables - size and diversity of jurisdictions, and the nature of the party system. On the first matter, the theory of populist federalism argues for smaller, more homogenous communities where the general will can be rnore clearly established. This runs counter to Madison's claim that size and diversity would help to protect fieedom. In the rights-based discourse of the late 20th century, marginalizing minority groups would be seen as anachronistic at best. It is not clear, however, that the use of direct democracy leads to this. The overall frequency of ballot measures among the 11 jurisdictions studied here declines as the size of the jurisdiction declines, and ballot measures threatening to minorities virtually disappear. This is not to say the threat to minorities is non-existent; it may come from legislatures or some other source. Rather, the argument here is that it cannot be clearly established that the use of populist direct democracy devices leads to increased harm, nor that it does so in smaller communities.

The nature of the party system also has an impact on the relationship between populism and federalism. Referendums are most likely to pass when they have all-party support, thus consensual systems have greater opportunities for passage of ballot 168 measures than do adversarial systems. The Australian example provides a stark reminder

of this. However, even within a consensual system, another element must be considered.

In all four countries, the party system is somewhat federalized, either with different

parties running at the national and sub-national levels, or with sub-national parties having

varying levels of autonomy fiom their federal masters. The presence of autonomous sub-

national parties brings more pIayers into the game and makes consensus more difficult to

reach. Cantonal parties in Switzerland have often been the base of opposition to

referendums and the founders of successful initiatives. The opposition of a few

provincial parties (and of regionally based federal parties) was important to the defeat of

the Charlottetown Accord. Such parties may not always have sufficient support to be

victorious, but they do make the negotiation process more difficult.

While the analysis focused on these two intervening variables, some others did

emerge. Most notably, it is possible the institutions of Parliamentary systems, including

the Crown, may have limited the growth of the most populist of direct democracy

devices, the initiative. Thus, the initiative found more fertile ground in Switzerland and the United States than in Canada and Australia.

4) How are the 'jpeople"defined?

This is the central theoretical question toward which all other questions are directed. Populism and federalism have been shown to co-exist, but how far the theoretical relationship can be taken is less clear. Populism seeks the general will of a monist people. Can this be attained in federal systems? 169 To a degree, yes. As a baseline, the presence of extensive ballot measure use in sub-national units in Switzerland and the United States demonstrates this. In fact, in the most extreme example, the handful of Landsgeminde in Swiss cantons indicate that in internal affairs, sub-national units can use processes extremely consistent with populism.

Limits to this communal liberty are present both through the demands that come with the monetary benefits of fiscal federalism, and by the instituting of Bills and Charters of

Rights that assume and impose universal standards. Even in the theoretical construct of a system with no recognition of universal rights and pure dual federalism, the people of the sub-national unit would still be yielding to a broader definition of the people on areas of national jurisdiction. In such a case, one body would constitute "the people" on some issues, another body would constitute "the people" on another. However, the presence of double majorities means that on issues where federalism is recognized to be important, such as constitutional amendments, both sets of "people" can be heard simultaneously.

The problem with the full co-existence of populism and federalism is that federalism does not make these neat lines of division. Federalism does divide power, but in practice the constitutional lines drawn are first of all, not neat, and second, not the last word. All federal constitutions have discussions of implied powers which become subject to interpretation by the courts and others. Cooperative federalism alters the lines even further, sometimes making them almost irrelevant. The lines of jurisdiction are not steadfast and static, they are malleable and in a state of flux. Interpretation of the changing nature of those lines often occurs through negotiation, negotiation between leaders and elites of the various levels of government. This sort of negotiation is problematic at best fiom a populist viewpoint. TOWARD A 'TOPULIST FEDERALISM"?

In chapter two, populism was said to be made up of two elements - a monist view of society and a belief in increased popular participation through direct democracy. Each of these elements have a different relationship to federalism.

In theoretical terms, the diversity of federalism cannot co-exist with monism. If one looks strictly at the national level of these four states, each recognizes diversity through various means, such as state or cantonal representation in the upper house, or perhaps through regional distribution of executive branch positions. At the sub-national level, however, monism is possible, although certainly not required. The Swiss case perhaps demonstrates this best. The Federal Council takes language, party and religion into account in its seven-person membership, and the upper house, the Standerat, has equal representation of each full canton. But the cantons are, historically at least, delineated on homogeneous lines, with most having one overwhelmingly dominant language and religion. Further, Swiss legislative and, especially, executive structures, work on a consensus basis that assumes nearly everyone can be brought to a common position.

The other aspect of populism, that of direct democracy, also has a complex relationship to federalism. It is not as inherently against federalism as monism can theoretically be seen to be, especially once the idea of the double majority is introduced for use at the national level. Only when one digs deeper into federalism and recognizes its tendencies toward elite negotiation does there appear to be a true conflict, but the relationship even then is more complex. Among federalism's characteristics is its 171 dividing of power. To put this in slightly different terms, federalism provides additional

entry points into the policy process. From this perspective, federalism and direct

democracy serve very similar roles. ~arei~is direct democracy used as the sole decision-

making process. About the only exception to that rule for the four countries in this study

would be in the three Swiss landsgemeinde cantons and some communes, and perhaps in

the town meetings of New England. Instead, the various types of ballot measures are used as either an additional entry point into the system or as a check against other parts of the political system. The referendum, in both the obligatory and facilitative form, is almost always used as the latter. The initiative is typically used as the former, although initiatives may be used in lieu of the facilitative referendum as a check on legislatures.

Such a use of direct democracy is most evident in Switzerland. It has been pointed out that the Swiss legislative system is one of elite consensus. Direct democracy provides a popular check on this system, meaning agreements made in the process of elite negotiation must have broad acceptance lest they be subject to a facilitative referendum.

The initiative also provides those not included in the consensus, or even those within it but seeing themselves as undemepresented on a given issue, an avenue to push issues onto the agenda. The double majority provision provides further protection, giving smaller cantons security that they will not be overwhelmed by the sheer numbers of the larger ones. Within some of the cantons, Neuchatel being the best example, fhther accountability of the legislative branch is provided through obligatory referendums on one-time spending over a certain level.

The pgssibility the Swiss experience is unique must be taken into account, but at least some of the same characteristics are present in the other three countries. In the United States, the goal of the populists at the beginning of the 1900s was clearly to

reform the abuses of a representative system they saw as corrupt, thereby opening up new

entry points into a system viewed as loaded against their interests. On the other hand,

there seems to be less indication than in Switzerland of using ballot measures as either a

consensual device or a check on representative action. Among the five states in this

study, the indirect initiative is available only in Washington, and has not been widely

used, with most initiative organizers preferring to go straight to the people through the

direct initiative. The facilitative referendum is available in all five states, but has seen

regular use only in North Dakota.

Australia and, especially, Canada, make less use of direct democracy, but the idea

of a check can still be seen; In Australia, elite negotiation alone cannot amend the

constitution. Amendments must be approved by the people, and the nature of the double

majority system in Australia, with four of six states needing to approve, makes it unusual

to succeed with a narrow majority of the populace. In the Canadian case, the referendum

on the Charlottetown Accord should be seen as a vote providing a popular check on a

process of elite accommodation. Further, the introduction of initiatives in two Canadian

provinces, even if not yet used, demonstrates the barriers of the Crown and responsible

government are not as high as they were when the first wave of direct democracy struck

in the early part of the 20' century.

To varying degrees, then, but to at least some degree in all four countries, direct

democracy and federalism serve complementary roles for either building consensus,

allowing new entry points into the policy process, or as a check on legislative behavior.

This, however, raises another theoretical question. If ballot measures are simply another 173 door into the pluralist room, how much can we associate them with their monist roots in

populism? One answer is that the idea of "the common sense of the common people" is

at its roots monist in that it assumes "common sense" is somehow unified. To the extent

direct democracy seeks the "common sense of the common people," even if that is merely one more input into the system, it retains a link to monism.

Can there be a populist federalism? The answer is a qualified yes. The two concepts can and do work together. In particular, the linking of direct democracy and federalism appears to slow down centrifugal movement of power to the center, allowing more to be done in smaller, more localized jurisdictions. Therefore, the federal bargain does provide some space for sub-national groups to define themselves as "people." But in the end, it fails to fully answer the question of just who the people are. Are the people who determine separation, whether that be in Quebec, Western Australia, or somewhere else, only those separating, or does a broader definition of people come into play? Even when a clear dividing line has been established, the answer is not entirely clear.

Federalism can never provide a perfect, all-encompassing answer for the question populist institutions must ask. Does this mean the two are intractable enemies? No. But like federalism itself, the definition of the people the relationship between populism and federalism enables is not all-encompassing. It, too, is malleable and fluctuating. BIBLIOGRAHPY

Allcock, J.B. (1971). "Populism: A Brief Biography," Sociology 5:371-87.

Almond, Gabriel & Sidney Verba (1965): The Civic Culture: Political Attitudes and Democracy in Five Nations, An Analytic Study (Boston: LittleBrown).

Argersinger, Peter (1974). Populism and Politics: WilliamAlfLed Pefler and the People's Party (Lexington: Kentucky).

Barber, Benjamin R. (1974). The Death of Communal Liberty: A History of Freedom in a Swiss Mountain Canton (Princeton, NJ: Princeton).

Barber, Benjamin R. (1 980). Strong Democracy (Berkeley: California).

Brutus [1787](1986). "Essay I", in Ralph Ketcham, ed. The Anti-Federalist Papers and the Constitutional Convention Debates (New York: Mentor).

Bowler, Sham & Todd Donovan (1998). Demanding Choices: Opinion, Voting, and Direct Democracy (Ann Arbor, MN:Michigan).

Boyer, Patrick (1992a). Direct Democracy in Canada: The History and Future of Referendums (Toronto: Dundurn).

Boyer, Patrick (1 992b). Mandate of the People: Referendums and a More Democratic Canada (Toronto: Dundurn).

Brugger, Bill & Dean Jaensch (1985). Australian Politics: neory and Practice (Sydney: Allen & Unwin).

Black, Conrad (1 977). Duplessis (Toronto: McClelland & Stewart).

Black, Ed* (1989). "Going Public: Mass Communications and Executive Federalism," in David P. Shugaman & Reg Whitaker (eds.), Federalism and Political Community: Essays in Honour of Donald Smiley (Peterbourough: Broadview).

Burgess, Michael (1993). "Federalism and Federation: A Reappraisal," in Michael Burgess & Alain-G. Gagnon, eds., Comparative Federalism and Federation: Competing Traditions and Future Directions (Toronto: U of Toronto).

Bustillio, Miguel(2000). "State GOP Is Distancing Itself From Prop. 187," Los Angeles Times, Aug. 7,2000.

Butler, David and Austin Ranney, eds. (1978). Referendums: A Comparative Study of Practice and Theory (Washington: AEI Press). Butler, David and Austin Ranney, eds. (1994). Referendums Around the World (Washingt0n:AEI).

Cairns, Alan (1 988). "The Governments and Societies of Canadian Federalism," in Constitution, Government and Society in Canada, Douglas Williams, ed. (Toronto: McClelland & Stewart).

Cairns, Alan (1992). "The Charter: A Political Science Perspective," Osgoode Hall Law Journal, Vol. 30, No.3.

Canadian Refoxm-Conservative Alliance Party (2000). "Your Principles: Declaration of Policy," www.canadianalliance.ca~yourprinciples/policy~declare/hdex.h~l.

Canovan, Margaret (1981). Populism (London: Junction Books).

Clark, Manning (1 963). A Short History of Australia (Carlion, Victoria: Melbourne).

Ciffe, Lionel (1967). One Party Democracy: The 1965 Tanzania General Elections (Nairobi: East African Pub. House).

Citrin, Jack (1984). "Introduction: The Legacy of Proposition 13," in Terry Schwadron, ed. California and the American Tax Revolt (Berkeley: California).

Crisp, L.F. (1983). Australian National Government, 6'" ed. (Melbourne: Longman Chesire).

Cronin, Thomas E. (1988). Direct Democracy: The Politics of Initiative, Referendum, and Recall (Cambridge, MA: Harvard).

Davis, S. Rufus (197811, The Federal Principle: A Journey Through Time in Quest of a Meaning (Berkeley: California).

Delley, Jean-Daniel & Andreas Auer (1986). "Structures politiques des cantons," in Rairnond E. Gemann & Ernst Weibel with Haans Peter Graf, eds. Manuel Systeme politque de la Suisse, Volume 3: Federalisme (Berne: Haupt).

Donovan, Todd and Shaun Bowler (1998). "An Overview of Direct Democracy in the American States," in Shaun Bowler, Todd Donovan & Caroline J. Tolbert, eds., Citizens as Legislators: Direct Democracy in the United States (Columbus: Ohio State).

Donovan, Todd and Sham Bowler (1998). "Direct Democracy and Minority Rights: An Extension," American Journal of Political Science 42: 1020-24. Duchacek, Ivo (1 970). Comparative Federalism: The Territorial Dimension of Politics (New York: Holt, Rinehart and Winston).

Else-Mitchell, R. (1982). "Constitutional Aspects of Commonwealth and State Taxing Laws," in Andrew Parkin, John Summers & Dennis Woodward, eds., Government and Politics in Australia, 2nd edition (Melbourne: Longman Chesire).

Elazar, Daniel J. (1972). American Federalism: A Viavfiom the States, 2nd edition (New York: Crowell).

Elazar, Daniel J. (1993). "Communal Democracy and Liberal Democracy," in Publius 23 (Spring 1993): 3-18.

Evans, Michael (1999). "MODsacking halted after gays win Forces case," The Times (London), Sept. 28.

Fiorina, Moms (1980). "Collective Responsibility in American Politics," Daedalus 109:3 (25-45).

Flanagan, Tom (1995). Waitingfor the Wave: The Reform Party and Preston Manning (Toronto: Stoddard).

Frenkel, Max (1986). Federal Theory (Canberra: Centre for Research on Financial Relations).

Frenkel, Max (1993). "The Communal Basis of Swiss Liberty," in Publius 23 (Spring 1993): 61-70,

Galligan, Brian (1989). 6cIntroduction,"in Brian Galligan, ed. Australian Federalism (Melbourne: Longman Cheshire).

Galligan, Brian (1 995). A Federal Republic (Melbourne: Cambridge).

Galligan, Brian & Cliff Walsh (1990). "Australian Federalism: Development and Prospects," in Publius 20-4 (Fall 1990): 1-17.

Gamble, Barbara S. (1997). "Putting Civil Rights to a Popular Vote," American Journal of Political Science 9 1:245-69.

Gibbins, Roger (1 990). Conflict and Unity: An Introduction to Canadian Political Life, 2nd ed., (Toronto: Methuen).

Gibbins, Roger and David Thomas (1993). "Ten Lessons From the Referendum,'' Canadian Parliamentary Review 4: 15 (Winter 1992-93): 3-6. Gollan, R.A.(1955). "Nationalism, the Labour Movement and the Commonwealth," in Gordon Greenwood, ed., Australia: A Social and Political History (New York: Praeger).

Greenberg, Edward and B. Page (1993). The Struggle for Democracy (New York: Addison Wesley).

Hauss, Charles (1994). Comparative Politics: Domestic Responses to Global Challenges, 2nd ed. (St. Paul: West).

Hughes, Colin (1 994). "Australia and New Zealand," in David Butler & Austin Ranney, eds., Referendums Around the World: The Growing Use of Direct Democracy (Washington: AEI Press).

Hughes, Rob (1999). "Brussels flexes muscles over players' rights, " The Times (London), Dec. 2.

Jackson, Robert & Doreen Jackson (1 992). Politics in Canada: Culture, Institutions, Behavior and Public Policy, 3rd edition (Scarborough, Ont.: Prentice-Hall Canada).

Jaensch, Dean (198 1). An Introduction to Australian Politics (Melbourne: Longman Cheshire).

Jaensch, Dean (1994). Politics of Australia (Melbourne: MacMillian).

Johnston, Richard, et a1 (1996). The Challenge of Direct Democracy: The 1992 Canadian Referendum (Montreal & Kingston: McGill-Queens).

King, Preston T. (1982). Federalism and Federation (Baltimore, Johns Hopkins).

Kobach, Kris W. (1993). Tize Referendum: Direct Democracy in Switzerland (Aldershot: Dartmouth).

Koenig, Louis (1971). Bryan: A Political Biography of William Jennings Bryan (New York: Putnam).

LaNauze, John A. (1972). The Making of the Australian Constitution (Melbourne: Melbourne).

Laycock, David (1990). Populism and Democratic Thought in the Canadian Prairies, 1910-1945 (Toronto: Toronto).

Lee, Eugene (1 978). "California," in David Butler and Austin Ranney, eds. Refeendums: A Comparative Study of Practice and Theory (Washington: AEI Press). Lijphart, Arend (1975). "The Comparable-Cases Strategy in Comparative Research," in Comparative Political Studies 8: 131-9.

Lijphart, Arend (1985). 'Won-Majoritarian Democracy: A Comparison of Federal and Consociational Theories," Publius 15 (Vo1.2):3-16.

Linder, Wolf (1995). Swiss Democracy: Possible Solutions to Conflict in Multicultural Societies (New York: St. Martin's).

Lipset, Seymour Martin (1981). Political Man: The Social Bases of Politics (Baltimore: Johns Hopkins).

Lipset, Seymour M. & Stein Rokkan (1967). Party Systems and Voter Alignments: Cross- National Perspectives (New York: Free Press).

Livingston, William L. (1956). Federalism and Constitutional Change (Oxford: Clarendon Press).

Loughlin, John (1986). "Federalist and Regionalist Movements in France," in Michael Burgess, ed. Federalism and Federation in Western Europe (London: Croom Helm).

Lucy, R. (1994). The Australian Form of Government, 2ndedition(Melbourne: Longman Chesire).

Publius (James Madison) [1788](1961). "Federalist #lo, #39, #49, #5 1," The Federalist Papers, Clinton Rossiter, ed. (New York: Penguin).

MacRae, Donald (1969). "Populism as an Ideology," in Ghita Ionescu & Ernest Gellner, eds. Populism: Its Meaning and National Characteristics (New York: Macrnillan).

Magleby, David B. (1 984). Direct Legislation: Voting on Ballot Propositions in the United States (Baltimore: Johns Hopkins).

Magleby, David B. (1998). "Ballot Initiatives and Intergovernmental Relations in the United States," Publius (Winter 1998):147-63.

Matas, Robert (1993). "B.C. eyes strict recall rules," Globe and Mail, Nov. 24, A1-2.

McDonnell, Patrick J. (1999). "Davis Won't Appeal Prop. 187 Ruling, Ending Court Battles," Los Angeles Times, July 29.

Michels, Roberto [I915](1962). Political Parties: A Sociological Study of the Oligarchical Tendencies of Modern Democracy (New York: Collier). Morton, W.L. (1944). "Direct legislation and the origins of the Progressive movement," Canadian Historical Review 25,2:279-88.

Morton, W.L. (1957). Manitoba: A History (Toronto: Toronto).

Norton, Phillip (1994). The British Polity, 3d edition (White Plains, NY: Longman).

Painter, Martin (1996). "The Council of Australian Governments and Intergovemental Relations: A Case of Cooperative Federalism," in Publius 26:2 (Spring 1996): 10 1-120.

Penniman, Howard, ed. (1 978). The Australian National Election of 1977 (Washington: AEI).

Price, Vincent (1992). Communications Concepts 4: Public Opinion (Newbury Park, CA: Sage).

Ranney, Austin, ed (1 98 1). The Referendum Device: A Conference (Washington: AEI).

Reform Party of Canada (1991). Reform Blue Book (Calgary).

Reform Party of Canada (1992). "Reform Party Caucus Issue Statement 36," July 16.

Riker, William H. (1964). Federalism: Origin, Operation, Significance (Boston: Little, Brown and Co.).

Riker, William H. (1982). Liberalism against Populism :a conzontation between the theory of democracy and the theory of social choice (Prospect Heights, Ill.: Waveland).

Rohr, Jean (1987). La democratie de Suisse (Paris:Economica).

Rousseau, Jean-Jacques [1762](1947). The Social Contract, Charles Frankel, ed. (New York: Hafner).

Saunders, Cheryl A. (1990). " Government Borrowing in Australia," in Publius 20-4 (Fall 1990): 35-52.

Scarrow, Susan E. (1999). "Democracy Within - And Without - Parties," Party Politics 5(3):275-282.

Shaman, Campbell (1999). "Taxation and Revenue Raising in the States," in Murray Good and James Gillespie, eds. Taxing Australia (Melbourne: Melbourne). Sirneon, Richard & Katherine Swinton (1995). "Introduction: Rethinking Federalism in a Changing World," Karen Kopp, et al, eds. Rethinking Federalism: Citizens, Markets, and Governments in a Changing World (Vancouver: UBC).

Smiley, Donald V. (1987). The Federal Condition in Canada (Toronto: McGraw-Hill Ryerson).

Steinberg, Jonathan (1 996). Why Switzerland?, 2nd edition (London: Cambridge).

Stevenson, Garth (1989). Unfilfilled Union: Canadian Federalism and National Unity (Agincourt, Ont. : Gage).

Storing, Herbert (1981). The Complete Anti-Federalist, Vol. 1 (Chicago).

Tolbert, Caroline J., Daniel H. Lowenstein & Todd Donovan (1998). "Election Law and Rules for Using Initiatives," in Shaun Bowler, Todd Donovan & Caroline J. Tolbert, eds., Citizens as Legislators: Direct Democracy in the United States (Columbus: Ohio State).

Tolley, Michael C. & Bruce A. Wallin (1995). "Coercive Federalism and the Search for Constitutional Limits," Publius 25:4 (Fall 1995):73-90.

Treschel, Alexandre H. & Hanspeter Kriesi, (1996). "Switzerland: the referendum and initiative as a centerpiece of a political system," in Michael Gallagher and Piervincenzo Uleri, eds., The Referendum Experience in Europe (LondonfNew York: MacMillanISt. James).

Trudeau, Pierre Elliott (1993). Memoirs (Toronto: McClelland and Stewart).

Verney, Douglas V. (1995). "Federalism, Federative Systems, and Federations: The United States, Canada, and India," Publius 25: 2 (Spring 1995): 8 1-97.

Wheare, K.C. (1964). Federal Government (New York: Oxford).

Wiltshire, Kenneth (1 992). "Australia's New Federalism: Recipes for Marble Cakes," in Publius 22-3 (Summer 1992): 165-180.

Wiles, Peter (1 969) "A Syndrome, Not a Doctrine: Some Elementary Theses on Populism" in Ghita Ionescu & Ernest Gellner, eds., Populism: Its Meanings and National Charcteristics (London: Weidenfield & Nicolson).

Witt, Stephanie and Suzanne McCorkle (1997) Anti-Gay Rights: Assessing Voter Initiatives (Westport, CT: Praeger). Zines, Leslie (1990), "Federal Constitutional Power Over the Economy," in PubIius 20-4 (Fall 1990): 18-34. APPENDIX 1

BALLOT MEASURES INVOLVING MAJORITY RIGHTS - 1970-94

UNITED STATES

MONTANA 1992 -Expansion Referendum - Increase membership on Board of Regents by with one being a Native American - FAILED

WASHINGTON 1978 - Threatening Initiative - Anti-busing - Prohibition on assigning students to other than nearest or next-nearest schools. PASSED

OREGON 1970 - Expansion Referendum - Repeal "White Foreigner" section of constitution. PASSED

1980 - Expansion Referendum - Guarantees voting rights to mentally handicapped unless incompetent. PASSED

1988 - Threatening Initiative - Revokes ban of sexual orientation discrimination in state executive branch. PASSED

1992 - Threatening Initiative - Government cannot facilitate, must discourage homosexuality and other 'behaviors' - FAILED

1994 - Threatening Initiative - Constitutional amendment that governments cannot approve of or create classifications based on homosexuality. - FAILED

CALIFORNIA 1970 - Expansion Referendum - Revision of range of constitutional items, including "discrimination based on gender." PASSED

1972 - Threatening Initiative - Assignment of students to schools (Anti-busing for immigration) FAILED

1972 - Expansion Referendum - Granting naturalized citizens voting eligibility. PASSED.

1978 - Threatening Initiative - Ban homosexuals from teaching in public schools. FAILED 1979 - Threatening Referendum - Assignment of students to schools (Anti-busing). PASSED.

1984 - Threatening Initiative - Voting materials printed in English only. PASSED 1986 - Threatening Initiative - Barred those with AIDS fiom teaching or working in food- related jobs. FAILED

1986 - Threatening Initiative - Declares English official state language. PASSED. 1988 - Threatening Initiative - AIDS declared communicable disease subject to quarantine. FAILED.

1988 - Threatening Initiative - Removes confidentiality guarantees fiom AIDS tests. FAILED.

SWITZERLAND

FEDERAL 1970 - Threatening initiative - Seeking reduction in number of foreigners - FAILED

1971 - Expansion referendum - Granting franchise to women - PASSED

1973 - Expansion referendum - Repeal of confessional articles - PASSED 1974 - Threatening initiative - Seeking reduction in number of foreigners - FAILED

1977 - Threatening initiative - Seeking reduction in number of foreigners - FAILED

1977 - Threatening initiative - Limits on naturalization of foreigners - FAILED

1979 - Expansion referendum - Lowering voting age to 18 - FAILED

1981 - Threatening initiative - Adjust policy towid foreign residents - FAILED

1981 - Expansion referendum - Equal rights for men and women - PASSED

1983 - Expansion referendum - Revise naturalization policy - PASSED

1983 - Expansion referendum - Ease naturalization rules - FAILED 1984 - Threatening initiative -No selling of land to foreigners - FAILED

1987 - Expansion referendum - Loosening of laws regarding asylum - PASSED 1987 - Expansion referendum - Revision of laws regarding foreigners - PASSED

1988 - Threatening initiative - Restrictions on immigration - FAILED

1991 - Expansian referendum - Reducing voting age to 18 - PASSED

1994 - Expansion referendum - Simplifying naturalization for young resident aliens - FAILED

1995 - Threatening referendum - Measures of constraint regarding the rights of foreigners (abuse of asylum) - PASSED

FRIBOURG 1970 - Expansion referendum - Extension of franchise to women -PASSED

1983 - Expansion referendum - Loweiing age of "activeyyvoting rights - PASSED 1991 - Expansion referendum - Lowering the voting age to 18 - PASSED

GENEVA 1972 - Expansion referendum - Lowering the age to exercise political rights - FAILED

1973 - Expansion referendum - Loosening of rules on nationality - PASSED

1980 - Expansion referendum - Lowering the age of majority to 18 years - PASSED

1987 - Expansion referendum - Guaranteeing equal rights for men and women - PASSED .

1993 - Expansion initiative - Granting voting rights to resident aliens - FAILED

LUZERN 1970 - Expansion referendum - Extension of hchise to women - PASSED

198 1 - Expansion referendum - Lowering of voting age to 18 - FAILED

1985 - Expansion initiative - "Equal education for boys and girls" - FAILED

1986 - Expansion referendum - Lowering commune voting age to 18 - PASSED

1992 - Expansion referendum - Lowering "activey' votbg age to 18 - PASSED NEUCHATEL 1979 - Expansion referendum - Lowering voting age to 18 - PASSED

1990 - Expansion referendum - Allowing resident aliens to hold office in commune governments - FAILED ZURICH 1970 - Expansion referendum - Extending the bchise to women - PASSED

1973 - Expansion referendum - Lowering the voting age - FAILED

1976 - Expansion referendum - Approval of annual loan used for educating refugees - FAILED

1980 - Expansion referendum - Lowering the voting age - PASSED 1993 - Expansion referendum - Granting right to vote in local elections to legal aliens - FAILED APPENDIX 2

CENTIUEIZING BALLOT MEASURES IN AUSTRALIA AND SWITZERLAND

SWITZERLAND (SINCE 1970)

1970 - Federal government to take role in encouraging sport and physical education. Popular vote: Yes '74.6%, No 25.4%; Cantons: Yes 22, No 0. PASSED.

1970 - Federal government to gain jurisdiction over environmental protection. Popular vote: Yes 92.7%, No 7.3%; Cantons: Yes 22, No 0. PASSED.

1973 - Initiative granting federal government control of education. Popular vote: Yes 52.8%, No 47.6%; Cantons: Yes 10.5, No 11.5. FAILED.

1973 - Urgent decree granting federal control of wages and prices. Popular vote: Yes 59.8, No 40.2; Cantons: Yes 20, No 2. PASSED.

1973 - Urgent decree granting federal control to limit credit. Popular vote: Yes 65.1%, 34.9%; Cantons: Yes 18.5, No 2.5. PASSED.

1976 - Urgent decree granting federal control of wages and prices. Popular vote: Yes 82.0%, No 18.0%; Cantons: Yes 22, No 0. PASSED

1977 - Introduction of a federal value added tax. Popular vote: Yes 40.5%, No 59.5%; Cantons: Yes 1, No 2 1. FAILED.

1979 - Introduction of a federal value added tax. Popular vote: Yes 34.6%, No 65.4%; Cantons: Yes 0, No 23. FAILED

1983 - Federal government to gain jurisdiction over energy policy. Popular vote: Yes 50.9%, No 49.1%; Cantons: Yes 1 1, No 12. FAILED.

1988 - Federal government to gain power to coordinate transportation policy. Popular vote: Yes 45.5%, No 54.5%; Cantons: Yes 3.5, No 19.5. FAILED.

1990 - Federal government to gain jurisdiction over energy policy. Popular vote: Yes 71.0%, No 29.0%; Cantons: Yes 23, No 0. PASSED

AUSTRALIA

1910 - Commonwealth to gain power over state debts. Popular Vote: Yes 54.9%, No 45.1%; States: Yes 5, No 1. PASSED. 191 1 - Commonwealth to gain legislative power over wide range of economic matters. Popular vote: Yes 39.4%, No 60.6%; States: Yes 1, No 5. FAILED.

191 1 - Commonwealth to gain power to nationalize monopolies. Popular vote: Yes 39.9%, No 60.1%; States: Yes 1, No 5.

1913 - Commonwealth to gain power over trade and commerce. Popular vote: Yes 48.8%,No 51.2%; States Yes 3, No 3.

1913 - ~oxnhonwealthto gain power over corporations. Popular vote: Yes 49.3%,No 50.7%;States: Yes 3, No 3. FAILED.

1913 - Commonwealth to gain power over industrial matters. Popular vote: Yes 49.3%, No 50.7%;States: Yes 3, No 3. FAILED.

1913 - Commonwealth to gain power over trusts. Popular vote: Yes 49.8%,No 50.2%; States: Yes 3, No 3. FAILED.

1913 - Commonwealth to gain power to nationalize monopolies. Popular vote: Yes 49.3%, No 50.7%; States: Yes 3, No 3. FAILED.

1913 - Commonwealth to gain jurisdiction over industrial disputes involving state railways. Popular vote: Yes 49.1%,No 50.9%; States: Yes 3, No 3. FAILED.

1919 - Commonwealth to gain temporary extension of wartime powers over wide range of economic matters. Popular vote: Yes: 49.7%,No 50.3%; States: Yes 3, No 3. FAILED.

191 9 - Commonwealth to gain power to nationalize monopolies. Popular vote: Yes 48.6%, No 5 1.4%; States: Yes 3, No 3. FAILED.

1926 - Commonwealth to gain authority to create controlling authorities over industry and commerce. Popular vote: Yes 43.5%,No 56.5%; States: Yes 2, No 4. FAILED.

1926 - Commonwealth to gain power to protect public fiom intemption of essential services. Popular vote: Yes 42.8%, No 57.2%; States: Yes 2, No 4. FAILED.

1928 - To put end to system of per capita payments to the states and restricting state borrowing by submitting it to a Loan Council. Popular vote: Yes 74.3%,No 25.7%; States: Yes 6, No 0. PASSED

1937 - Commonwealth to gain power to legislate on aviation. Popular vote: Yes 53.6%, No 46.4%; States: Yes 2, No 4. FAILED. 1937 - Commonwealth to gain power over marketing. Popular vote: Yes 36.3%,No

, 63.7%; States: Yes 0, No 6. FAILED. 1944 - Commonwealth to gain jurisdiction over 14 matters for five-year post-war reconstruction period. Popular vote: Yes 46.0%, No 54.0°h; States: Yes 2, No 4. FAILED.

1946 - Commonwealth to gain power to legislate on social services. Popular vote: Yes 54.4%, No 45.6%; States: Yes 6, No 0. PASSED.

1946 - Commonwealth to gain lawmaking power over marketing boards. Popular vote: Yes 50.6%, No 49.4%. States: Yes 3, No 3. FAILED.

1946 - Commonwealth to gain power to legislate on tenns and conditions of industrial employment. Popular vote: Yes 50.3%, No 49.7%; States: Yes 3, No 3. FAILED.

1948 - Commonwealth to gain permanent power to control rents and prices. Popular vote: Yes 40.7%, No 59.3%. States: Yes 0, No 6. FAILED.

1951 - Commonwealth to gain power to deal with ~oxnhunisrn.Popular vote: Yes 49.4%, No 50.6%. States: Yes 3, No 3. FAILED.

1967 - Commonwealth to gain jurisdiction over Aboriginal affairs. Popular vote: Yes 90.8%, No 9.2%; States: Yes 6, No 0. PASSED

1973 - Commonwealth to gain control over prices. Popular vote: Yes 43.8%, No 56.2; States: Yes 0; No 6. FAILED

2973 - Commonwealth to gain control over incomes. Popular vote: Yes 34.4%, No 63.6%; States: Yes 0, No 6. FAILED

1974 - Commonwealth to gain power to borrow money on behalf of local government bodies. Popular vote: Yes 46.9%, No 53.1%. States: Yes 1 (New South Wales), No 5. FAILED.

1977 - To enable Commonwealth and states to voluntarily refer powers to one another. Popular vote: Yes 43.9%, No 56.1%; States: Yes 0, No 6. FAKED.